text
stringlengths 2
1.05M
| repo_name
stringlengths 5
101
| path
stringlengths 4
991
| language
stringclasses 3
values | license
stringclasses 5
values | size
int64 2
1.05M
|
|---|---|---|---|---|---|
I am the biggest Izzy hater on here, but we need to hold off on judgement
| null |
minipile
|
NaturalLanguage
|
mit
| null |
brandy cherries
June 29, 2011
by Deborah Weiler (Arlington, VT)
I love this for gifts. It is a sure hit.
A nice change from the usual cordials.
It gets people's attention.
Fills your mouth with goodness!
January 19, 2011
by Janice Tocher (Knoxville, TN)
I had bought this item with every good intention of giving as a Christmas gift. However, since it is now nearly a month past Christmas and it was sitting there, decided the gift was meant for me after all!
I popped one of these gems in my mouth and oh... my! The combination of flavor sensations was excellent. These are not your standard Chocolate Covered Cherries and I definitely would not recommend for those under drinking age. The premium brandy is just that... premium!
Of note: I taste tested the cherry about a half-hr ago and true to truly fine confections, the flavor is still dancing in my mouth - in a good way :)
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Right now at ShopBop you can check out a the James Jeans section, made delectable by a beautifully styled Autumn lookbook that’s been broken down by items you can buy elsewhere on the site. The actual shop selection is nicely varied, from knee-length denim skirts to skinny legging-jeans, boot-cuts, and even shorts. Check out some images from the look book below, and shop the full set of James Jeans here.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Our research program integrates functional studies with extensive structural analysis of pattern recognition receptors (PRRs) in complex with their ligands as well as downstream adapters and effector molecules. Both membrane-bound (such as the Toll-like receptors) and cytoplasmic (such as the CATERPILLAR family members and RIG-I/MDA5 antiviral proteins) PRRs are targeted. A critical feature of these innate immune receptors is that they distinguish among various classes of pathogen-specific molecules while retaining responsiveness to a large number of related molecules within a given biochemical class. Understanding such [unreadable][unreadable]Sbroad[unreadable][unreadable]? reactivity at the atomic level is one of the primary goals of the program. Ligand binding by the PRRs not only initiates intracellular signaling cascades leading to innate immune responses against infections, but also allows the innate immune system to orchestrate and potentiate the activities of the adaptive immune system. Our research aims to decipher this signaling network by studying structures of macromolecular complexes using x-ray crystallography.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
---
abstract: 'A geometric potential $V_C$ depending on the mean and Gaussian curvatures of a surface $\Sigma$ arises when confining a particle initially in a three-dimensional space $\Omega$ onto $\Sigma$ when the particle Hamiltonian $H_\Omega$ is taken proportional to the Laplacian $L$ on $\Omega$. In this work rather than assume $H_\Omega \propto L$, momenta $P_\eta$ Hermitian over $\Omega$ are constructed and used to derive an alternate Hamiltonian $H_\eta$. The procedure leading to $V_C$, when performed with $H_\eta$, is shown to yield $V_C = 0$. To obtain a measure of the difference between the two approaches, numerical results are presented for a toroidal model.'
author:
- 'M. Encinosa'
bibliography:
- 'hermbib.bib'
---
Introduction
============
The study of quantum mechanics on curved surfaces has been a subject of theoretical effort for decades [@jenskoppe; @dacosta1; @dacosta2; @exnerseba; @matsutani; @burgsjens; @duclosexner; @bindscatt; @goldjaffe; @ouyang; @popov; @midgwang; @clarbrac; @schujaff; @ee1; @ee2; @lin; @chapblic; @qu; @nils]. The approach as pioneered in [@burgsjens; @jenskoppe; @dacosta1; @dacosta2] first defines a surface $\Sigma(u,v)$ and employs a coordinate $q$ to describe excursions at any point normal to $\Sigma$. The Hamiltonian appropriate to the three dimensional coordinate system $\Omega$ characterized by a metric $g_{ij}$ is then defined by $H_\Omega
\equiv -{1\over2} \nabla^2$ with$$\nabla^2= g^{-{1 \over 2}}{\partial \over \partial q^i} \bigg [
g^{1 \over 2}\ g^{ij}{\partial \over \partial q^j} \bigg ].
\eqno(1)$$ As will be briefly reviewed below, Eq.(1), along with imposing conservation of the norm in the limiting case where the particle is constrained to motion on $\Sigma$, leads to a geometric potential $V_C$. This paper with is concerned with how $V_C$ is modified when instead of defining $H_\Omega$ through Eq.(1), another choice of Hamiltonian, one for which surface and normal momenta are required to be Hermitian over $\Omega$ is adopted.
The remainder of this paper is organized as follows: in section 2 the geometric potential is derived from Eq.(1) for a cylindrically symmetric geometry $\Omega_c$. This material has appeared other places, but is included here to keep this work as self contained as possible and to establish some notation. A symmetric geometry was chosen for ease of presentation; the salient points for more general cases remain intact. In section 3 the classical Hamiltonian for a particle in $\Omega_c$ is written, Hermitian momenta on $\Omega_c$ are found, and standard quantum mechanical rules employed to write the first quantized Hamiltonian. Finally, using the limiting procedure detailed in section 2, the geometric potential is re-derived. Section 4 presents numerical calculations for a toroidal system, and section 5 is reserved for conclusions.
The geometric potential; a brief review.
========================================
The derivation of $V_C$ will be performed for a cylindrically symmetric geometry $\Omega_c$ with $H_\Omega \propto L$. The extension to the general case is straightforward and the central results below will be not be affected by the restriction to azimuthal symmetry.
Let ${\mathbf e}_{\rho},{\mathbf e}_{\phi}$ be cylindrical coordinate system unit vectors defined through $$\rm {\mathbf e}_{\rho}=cos\phi \ {\mathbf i}+ sin \phi \ {\mathbf
j} \eqno(2)$$ $$\rm {\mathbf e}_{\phi}=-sin \phi \ {\mathbf i}+ cos \phi \
{\mathbf j} \eqno(3)$$ and let a cylindrically symmetric surface $\Sigma_c$ be described by $${\mathbf r}(\rho,\phi)=\rho {\mathbf e}_{\rho} + S(\rho)
{\mathbf k}. \eqno(4)$$ $S(\rho)$ gives the shape of the surface. Any point near $\Sigma_c$ may be reached by travelling normal to $\Sigma_c$ up (or down) a distance $q$ via $${\mathbf x}(\rho,\phi,q)={\mathbf r}(\rho,\phi)+q {\mathbf e}_{n}
\eqno(5)$$ with ${\mathbf e}_{n}$ everywhere normal to the surface. The metric near the surface derived from Eq. (5) is [@ee1] (subscripts here indicate differentiation) $$dx^2= Z^2 \bigg[ 1-{{qS_{\rho\rho}} \over {Z^3}} \bigg ]^2
d\rho^2 + \rho^2 \bigg[ 1-{{qS_{\rho}} \over {\rho Z}} \bigg ]^2
d\phi^2 +\ dq^2 \eqno(6)$$ $$\equiv Z^2 [ 1+q k_1 ]^2 d\rho^2 + \rho^2 [ 1+qk_2]^2 d\phi^2 +\
dq^2 \eqno(7)$$ with $Z=\sqrt {1+S_\rho^2}$. The Laplacian can be found straightforwardly from Eq. (7) but is not essential to the main arguments that follow.
Now consider a situation where a large confining potential $V_n(q)$ everywhere normal to $\Sigma_c$ acts to restrict the particle to the surface. The derivation of the geometric potential is independent of its specific form [@scripta], but however chosen it causes the particle to sit on $\Sigma_c$. In this limit the wave function decouples into surface and normal parts [@jenskoppe; @dacosta1; @dacosta2; @burgsjens] $$\Psi(\rho,\phi,q) \rightarrow \chi_s(\rho,\phi)\chi_n(q). \eqno(8)$$ Conservation of the norm in the decoupled limit is imposed by insisting upon the condition $$|\Psi|^2 F d\Sigma dq=|\chi_s|^2|\chi_n|^2d\Sigma dq \eqno(9)$$ with $F=1+2qh+q^2k$ and $d\Sigma$ the surface measure. $h$ and $k$ are the mean and Gaussian curvatures given by $$h={1\over 2}(k_1+k_2), \eqno(10)$$ $$k=k_1k_2. \eqno(11)$$ Inserting the expression given by Eq.(8) $$\Psi = {\chi_s\chi_n \over \sqrt F}
\eqno(12)$$ into the time independent Schrodinger equation $-{1\over 2}\nabla^2\Psi = E\Psi$, and taking the $q\rightarrow 0$ limit after performing all $q$ differentiations, yields the pair of equations $(\hbar = m = 1)$ $$-{1\over 2}\bigg[
{1\over Z^2}{\partial^2 \over \partial \rho^2}
+{1\over Z^2\rho}{\partial \over \partial \rho}
-{Z_{\rho}\over Z^3}{\partial \over \partial \rho}
+{1\over \rho^2}{\partial^2 \over \partial \phi^2}
+(h^2-k)\bigg]\chi_s =E_s
\chi_s \eqno(13)$$ $$-{1 \over 2}{\partial^2 \chi_n \over \partial q^2} + V_n(q)
\chi_n =E_n \chi_n. \eqno(14)$$ The curvature potential $V_C$ is $$V_C = -{1 \over 2}[h^2-k]; \eqno(15)$$ seeing specifically how $V_C$ is generated will prove important in the following section. It arises from the $q \rightarrow 0$ limit of the $$\bigg ( {\partial^2 \over \partial q^2} + 2h {\partial \over
\partial q}\bigg )
\chi_s(\rho,\phi)\chi_n(q) F^{-{1/2}} \eqno(16)$$ term in the full Laplacian . The $Z$ dependence appearing in the kinetic energy operator given by Eq. (13) occurs independently of the $q$ degree of freedom. The effect on the spectra of some simple systems due to metric factors in the operator was shown to be small in [@ee1; @ee2] for some parameterizations of $S(\rho)$, but it is also possible for the effects to be large [@encunpub].
The Hamiltonian with Hermitian momenta
======================================
In this section a Hamiltonian $H_\eta$ quadratic in momenta $P_\eta$ Hermitian over $\Omega_c$ is developed. The procedure followed comprises standard quantum mechanical rules, to wit:
a\. the classical Lagrangian is written b. the canonical momenta are determined c. the classical momenta are replaced with their first quantized form $\hat{P}_\eta$ Hermitian on $\Omega_c$.
The Lagrangian for a particle in a geometry described by Eq. (6) is $$L = {1\over 2} \big(Z^2[1+k_1q]^2 \dot{\rho}^2+ \rho^2[1+k_2q]^2
\dot{\phi}^2+\dot{q}^2 \big). \eqno(17)$$ With $P_\eta = {\partial L / \partial \dot{q}_\eta}$, the classical Hamiltonian for this geometry is $$H = {1\over 2} \bigg( {P_\rho^2 \over Z^2[1+k_1q]^2} + {P_\phi^2
\over \rho^2[1+k_2q]^2} + P_q^2 \bigg). \eqno(18)$$ To proceed with first quantization via $P_\eta \rightarrow
\hat{P}_\eta$, it is necessary to produce momentum operators Hermitian on a geometry described by $g_{ij}$. The required relation is $$\hat{P_\eta}={1\over i} \bigg( {\partial \over \partial q_\eta} +
{1 \over 2}{\partial \over \partial q_\eta}{\rm ln} \sqrt g
\bigg). \eqno(19)$$ Anticipating the $q \rightarrow 0$ limit, it is useful to note the differentiations done by the surface momenta will pass over any appearances of the $q$ variable, making it possible to set $q = 0$ in both of them without affecting the final result. The momenta in this limit are $$\hat{P}_\rho={1\over i} \bigg( {\partial \over
\partial \rho} + {1 \over 2}{\partial Z \over \partial \rho} + {1 \over 2\rho}
\bigg)\eqno(20)$$ $$\hat{P}_\phi={1\over i} {\partial \over
\partial \phi} \eqno(21)$$ $$\hat{P}_q={1\over i} \bigg( {\partial \over
\partial q} + {1 \over F}[h+qk]
\bigg). \eqno(22)$$ $H_\eta$ is straightforward to obtain from the three momenta listed above, but it is Eqs. (16) and (22) that will be focused upon to generate the central result of this paper. Write $$-\hat{P}_q^2= {\partial^2 \over \partial q^2}+ {2\over F}
(h+qk){\partial \over\partial q}+ {1 \over F} \bigg( {\partial h
\over \partial q} +k \bigg)- {1\over F^2} {\partial F \over
\partial q} (h+qk) + {1\over F^2} (h+qk)^2;
\eqno(23)$$ upon taking the $q \rightarrow 0$ limit the result is $$-\hat{P}_q^2= {\partial^2 \over \partial q^2}+ 2h {\partial
\over\partial q}+ k - h^2.
\eqno(24)$$ The key point is simple; Eq.(24) includes the operator that appears in the Laplacian formulation of the problem as shown in Eq.(16). It eventually yields $\partial^2 \over \partial q^2$ plus $h^2 - k$, from which it is trivially seen that the geometric potential cancels in the Hermitian formulation.
A numerical example
===================
In this section the spectra and wave functions of the Schrodinger equation for a particle on a toroidal surface $T^2$ of major radius $R$ and minor radius $a$ will be compared using the two formalisms detailed above. There are good reasons to choose such a system, not the least being calculating system observables is tractable. Additionally, toroidal nanostructures of varied types have been fabricated so there is some relevance to real devices, a point which will be elaborated upon in the conclusions section below.
The surface of a doughnut shaped torus has may be parameterized by [@fpl] $${\mathbf x}(\theta,\phi,q)=(R + a \ {\rm cos} \theta ){\mathbf e}_{\rho} +a\ {\rm sin}
\theta{\mathbf {k}} + q {\mathbf e}_{n}. \eqno(25)$$ Applying $d$ to Eq.(25) gives $$d{\mathbf x}= (a+q)d\theta {\mathbf e}_{\theta}+[R + (a+q) \ {\rm
cos} \theta]d\phi {\mathbf e}_{\phi}+dq {\mathbf e}_n \eqno(26)$$ with ${\mathbf e}_{\theta} =-{\rm sin} \theta {\mathbf
e}_{\rho}+{\rm cos}\theta {\mathbf k}$ and ${\mathbf e}_{n}\equiv
{\bf e}_{\phi} \ {\rm x} \ {\mathbf e}_{\theta}$.
Let $\alpha = a/R$, $\beta = 2Ea^2$; applying the formalism described in section 2 and making the standard separation ansatz for the azimuthal part of the wave function $\Psi(\theta,\phi) =
\psi(\theta)exp \ [i\nu\phi]$ yields the Schrodinger equation with $V_C$ present [@encmott], $${\partial^2 \psi \over \partial \theta^2} -
{\alpha \ {\rm sin}\ \theta \over [1 + \alpha \ \cos\theta]}{\partial \psi \over \partial \theta}
-{(\nu^2 \alpha^2- {1\over4}) \over [1 + \alpha \
\cos\theta]^2}\psi +\beta\psi = 0. \eqno(27)$$ The method of section 3 gives the Schrodinger equation $${\partial^2 \psi \over \partial \theta^2} -
{\alpha \ {\rm sin}\ \theta \over [1 + \alpha \ \cos\theta]}{\partial \psi \over \partial \theta}
-{[\nu^2 \alpha^2+ {1 \over 4}(\alpha^2 - 1)] \over [1 + \alpha \
\cos\theta]^2}\psi +(\beta-{1\over 4}\psi) = 0. \eqno(28)$$ Even before consideration of the solutions of Eq. (27) and (28), there is an immediate distinction between the two. The cancellation of the azimuthal kinetic energy term at discrete values of $\alpha$ occurs in the former at values of $$\alpha = {1 \over 2\nu} \eqno(29)$$ and in the latter at $$\alpha = \sqrt{{1 \over 1+4\nu^2}}. \eqno(30)$$ The ramifications of Eq.(29) and by implication Eq.(30) have been discussed elsewhere. Here it need be noted that while there appears a term in the Hermitian formalism that is identical to $V_C$ derived from the Laplacian method, it is an artifact of the high degree of symmetry of the torus and may take a very different form in the general case.
Numerical solutions of Eqs. (27) and (28) were determined by a basis set expansion in Gram-Schmidt functions on $T^2$ [@encgs]. Eigenvalues and wave functions are given in tables 1-3 for several low-lying states with $\alpha=1/3, 1/2$ and $2/3$. The results indicate clearly that the two Hamiltonians can differ, particularly at the magic" $\alpha$ given by Eq. (29) or Eq. (30), and when the torus fattens to values of $\alpha
\thicksim 2/3$.
Conclusions
===========
This work demonstrated that a Hamiltonian quadratic in momenta Hermitian on a curved surface and consistent with standard quantum mechanical rules leads to a vanishing geometric potential $V_C$. It is interesting that Golovnev has recently employed an alternate (Dirac quantization) procedure for which $V_C =0$ on a sphere [@golovnev]; however, it is not clear to the author how the DQ method could be easily implemented on $T^2$. Numerical results were presented to show the differences in spectra and wave functions between the two Hamiltonians discussed here can be substantial. It is possible to speculate as to whether advances in nanostructure fabrication may perhaps allow for experimental determination of which (if either) Hamiltonian studied here is the more correct description of curved surface nanophysics. Given that toroidal structures have been fabricated and are clean" with respect to curvature effects (in that there is enough symmetry to make the problem interesting but not so much to make it trivial), it is conceivable that existing devices may be employed to settle the very fundamental issue of what is the appropriate Hamiltonian for a particle constrained to a surface.
Acknowledgements
================
The author would like to thank Mr. Lonnie Mott for useful discussions.
$\quad \ \beta$ $\alpha = {1 \over 3}$ $V_C \neq 0$
------------------- ---------------------------------------------------------- --
-.2834 $\Psi_{0}=.4082- 0.0776\rm cos \theta$
-.1528 $\ \Psi_{1}=[.4049- 0.0421\rm cos \theta]e^{\pm i\phi}$
+.1968 $\ \Psi_{2}=[.4060- 0.0525\rm cos \theta]e^{\pm 2i\phi}$
$\quad \ \beta$ $\alpha = {1 \over 3}$ $V_C = 0$
0 $\ \Psi_{0}=-.4075- 0.0686\rm cos \theta$
+.1267 $\ \Psi_{1}=[.4038- 0.0335\rm cos \theta]e^{\pm i\phi}$
+.4735 $\ \Psi_{2}=[.3869- 0.0593\rm cos \theta]e^{\pm 2i\phi}$
: Eigenvalues and wave functions corresponding to the solutions of Eqs. (27) and (28) for $\alpha ={1 \over 3}$ (the $(2\pi)^{-{1\over2}}$ normalization from from the $\phi$ dependence is omitted.) Terms not shown are at least an order of magnitude smaller than those listed.
$\quad \ \beta$ $\alpha = {1 \over 2}$ $V_C \neq 0$
------------------- ----------------------------------------------------------- --
-.3511 $\Psi_{0}=.4230- 0.1470\rm cos \theta$
0 $\ \Psi_{1}= e^{\pm i\phi}$
+.6288 $\ \Psi_{2}=[-.3293- 0.1935\rm cos \theta]e^{\pm 2i\phi}$
$\quad \ \beta$ $\alpha = {1 \over 2}$ $V_C = 0$
0 $\ \Psi_{0}=.4191- 0.1072\rm cos \theta$
+.3192 $\ \Psi_{1}=[.3912+0.0293\rm cos \theta]e^{\pm i\phi}$
+.9212 $\ \Psi_{2}=[-.3289- 0.2008\rm cos \theta]e^{\pm 2i\phi}$
: As per table 1 with $\alpha ={1 \over 2}$. The middle state is qualitatively very different due to $\alpha ={1 \over 2}$ defining a magic radius" when $V_C = 0$.
$\quad \ \beta$ $\alpha = {2 \over 3}$ $V_C \neq 0$
------------------- ----------------------------------------------------------------- --
-.5947 $\Psi_{0}=0.4597- 0.3406\rm cos \theta-0.1000\rm cos 2 \theta$
+.2024 $\ \Psi_{1}= [0.3570+0.1079\rm cos \theta]e^{\pm i\phi}$
+.4377 $\ \Psi_{2}=0.5930\rm cos \theta-0.1363cos 2\theta$
$\quad \ \beta$ $\alpha = {2 \over 3}$ $V_C = 0$
3.4$(10^{-4})$ $\Psi_{0}=-0.4369+ 0.1516\rm cos \theta$
+.5397 $\ \Psi_{1}= [0.3405+0.1446\rm cos \theta]e^{\pm i\phi}$
+1.0000 $\ \Psi_{2}=-0.5888\rm sin \theta-0.0992 sin 2\theta$
: As per table 1 with $\alpha ={2 \over 3}$. The second excited state for the case when $V_C = 0$ becomes a negative parity function in $\theta$.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Is bitcoin an environmental monster? The energy consumption for production and entertainment however is enormous…
Bitcoin is getting more and more expensive. Recently, it noted with $ 7900 at a new ATH.
In the past ten days, a plus of 23 % is recorded!
The great interest in the digital currency also leads to the fact that previously overlooked aspects come into the spotlight.
For example, the cryptocurrency-focused internet platform Digiconomist calculates how much power the Bitcoin network consumes - and that's not all.
As it is mentioned, the currency is created digitally - mostly in gigantic mining farms (Proof of Work). In addition, the transactions are written to a file called the blockchain.
That's about 300,000 transactions per day. For their reward, the miner receive bitcoins.
As much electricity as Ireland
Running these computers consumes electricity, a lot of electricity. They generate bitcoin without break and that's why most of the data centers are located in China because each transaction requires around 245 kWh. That's about 3285 hours of television. That's 70.5 million kW hours a day 25.76 TWh a year. According to the average German electricity price, a Bitcoin transaction would cost more than € 70, in China it is only about € 23.
That is more than the country Nigeria consumes with 185 million inhabitants a year and about 0.12% of the global electricity consumption. And whoever argues that the power supply in Nigeria is underdeveloped: It is approximately the power consumption of Ireland.
Environmental consequences are controversial
The more transactions are made, the higher the power consumption. But it becomes difficult with the question whether Bitcoin is an environmental polluter - this depends on the current usage. There are data centers that are run with hydropower or geothermal energy.
However, since most server farms are located in China, we should assume that many data centers are running on coal. But you have to relativize the amount of electricity consumption: Even bills and coins are energy-intensive in the production. And even computer centers of the banks are not running with air and love. Everything that is managed digitally needs power.
The environmental consequences have been arguing for more than four years after author Mark Gimein published an article about it.
Insights into the high energy consumption of mining bitcoins is nothing new.
Already in mid-2015, a journalist had calculated that a single bitcoin transaction consumed as much energy as 1.7 US households a day. By the way, today there are already 8.3 households.
The Dutch scientist Sebastian Deetman had calculated scenarios for the energy consumption of bitcoins in spring 2016. The results for 2020 were between the output of a small power plant and the consumption of Denmark. Denmark currently consumes 33.3 TWh, 30% more than bitcoins.
So there is another solution to save unnecessary wasted energy…
Proof of Stake! The advantages of Proof of Stake are clear and there are several coins running on this algorithm. Key advantage is that Proof of Stake is a lot more energy efficient than Proof of Work as it can reduce the level of complex calculations and goes work involved that is inherent to it.
Proof of Stake only means people holding the currency earning it by verifying that they own. One famous Proof of Stake coin is Dash with its disadvantage of a quorum of 1000 Dash. You need to hold this amount be a Masternode before realizing any reward.
The alternative solution to Dash is Cloak which does not need more than one Cloak in order to get any staking rewards. The current team has really risen up from ashes and gave to the community a deep and clear understanding to aspect of privacy and anonymity. Cloakcoin provides earning interest of 6% per year by holding the currency.
Proof of Stake is important for the future considering the rise of mass energy consumption. It is a more reliable and ecological algorithm than the Bitcoin system. From the financial point of view, provides Proof of Stake also an advantage because no financial institution is rewarding any customer with a interest of 6% per year.
I hope with my article you will have a clear and deep understanding about the importance of Proof of Stake…
Thank you
Source: http://www.faz.net/aktuell/finanzen/digital-bezahlen/eine-bitcoin-transaktion-kostet-30-euro-strom-15282063.html
| null |
minipile
|
NaturalLanguage
|
mit
| null |
---
abstract: 'Existing automatic 3D image segmentation methods usually fail to meet the clinic use. Many studies have explored an interactive strategy to improve the image segmentation performance by iteratively incorporating user hints. However, the dynamic process for successive interactions is largely ignored. We here propose to model the dynamic process of iterative interactive image segmentation as a Markov decision process (MDP) and solve it with reinforcement learning (RL). Unfortunately, it is intractable to use single-agent RL for voxel-wise prediction due to the large exploration space. To reduce the exploration space to a tractable size, we treat each voxel as an agent with a shared voxel-level behavior strategy so that it can be solved with multi-agent reinforcement learning. An additional advantage of this multi-agent model is to capture the dependency among voxels for segmentation task. Meanwhile, to enrich the information of previous segmentations, we reserve the prediction uncertainty in the state space of MDP and derive an adjustment action space leading to a more precise and finer segmentation. In addition, to improve the efficiency of exploration, we design a relative cross-entropy gain-based reward to update the policy in a constrained direction. Experimental results on various medical datasets have shown that our method significantly outperforms existing state-of-the-art methods, with the advantage of fewer interactions and a faster convergence.'
author:
- |
Xuan Liao^1^ , Wenhao Li^2^[^1] , Qisen Xu^2^, Xiangfeng Wang^2^ ,\
Bo Jin^2^ , Xiaoyun Zhang^1^ , Ya Zhang^1^ , and Yanfeng Wang^1^\
^1^Shanghai Jiao Tong University , ^2^East China Normal University
bibliography:
- 'egbib.bib'
title: |
Iteratively-Refined Interactive 3D Medical Image Segmentation\
with Multi-Agent Reinforcement Learning
---
Introduction
============
Medical image segmentation has been widely recognized as an essential procedure for subsequent medical image processes such as structural and functional analysis, diagnosis and treatment. The traditional dense manual annotation is extremely inefficient for 3D medical images and its performance highly depends on the physician’s experience. With the development of convolutional neural networks (CNNs), automatic segmentation has greatly improved the efficiency of medical image process [@milletari2016v; @ronneberger2015u; @zhou2018unet++]. However, the accuracy and robustness for the current automatic methods still need to be improved for practical clinic use.
To get a better segmentation, interactive image segmentation [@bredell2018iterative; @cciccek20163d; @wang2018deepigeos] is introduced to integrate user hints (mostly in the form of points, scribbles and bounding boxes). This kind of interactive methods has become a popular research direction because it improves the performance of segmentation by adding new labeling constraints to the prediction model. Normally, a one-time interaction might not ensure the segmentation accuracy. Therefore, many existing methods are compatible with the iteratively-refined mode: the operator provides new hints according to the current result to refine the segmentation until it is satisfactory. Moreover, to reduce the number of interactions, the existing works introduce the idea of replacing the initial hints with an automatically-obtained coarse segmentation [@bredell2018iterative; @wang2018deepigeos]. Note that in this paper, we refer to such methods incorporating a coarse segmentation in the initial input as ***update methods*** and we will focus on this kind of methods in this paper.
Concerning the current update methods, there exist two main issues: 1) They usually ignore the dynamic process for successive interactions. Although the segmentation can be iteratively refined, the model always treat the segmentation of each refinement step in isolation, with an absence of the previous information. 2) Another problem is the loss of prediction uncertainty when using the binary segmentation result, instead of a segmentation probability for each voxel, as part of the model input. The rounding from dense segmentation probability to binary segmentation prediction may cause quantization error and accuracy loss.
To tackle the above two issues, this paper proposes a novel interactive medical image segmentation update method called *Iteratively-Refined interactive 3D medical image segmentation via Multi-agent Reinforcement Learning (IteR-MRL)*. We formulate the dynamic process of iterative interactive image segmentation as an MDP. Specifically, at each refinement step, the model needs to decide the labels of all voxels, according to the previous segmentations and supervision information from the interaction. After that, the model will get the feedback according to pre-defined measurement of segmentation, and the above process will be repeated until the maximum number of interactions is reached. We then adopt the RL methods to solve above MDP, that is, to find the segmentation strategy to maximize the accumulated feedbacks received at each refinement step. However, it will be intractable to use single-agent RL for voxel-wise prediction due to the large exploration space. In addition, considering that the voxels in the segmentation task are interdependent, they can achieve better segmentation by a more comprehensive grasp of the surrounding information. To reduce the exploration space to a tractable size and explicitly model the dependencies between voxels, we introduce the multi-agent reinforcement learning (MARL) method. We treat each voxel as an agent which decides its own label. All agents share the same policy and collaborate with each other through convolutional kernels. Meanwhile, instead of considering the difference between the current prediction and the ground truth, we design a relative cross-entropy gain-based reward to prompt agents to explore more efficiently. Specifically, the algorithm gives a positive reward for an improvement and vice versa at each refinement step, so that the new prediction can be forced to outperform the previous one. Compared with supervised methods, such RL-based methods have the advantage of a faster refinement convergence. The problem of prediction uncertainty loss in existing works caused by segmentation map binarization can be settled by adopting segmentation probability rather than binary segmentation as part of the RL state. This augmented state space also derives an adjustment action space leading to a more precise and finer segmentation. Then the segmentation refinement procedure can be regarded as a series of actions to adjust the segmentation probability with a certain level. In this way, the prediction uncertainty is reserved and the algorithm explores in a finer granularity and a denser space.
The experimental results indicate that the proposed IteR-MRL is robust to different initial segmentations and various medical datasets. Given the same initial segmentations, our proposed interactive algorithm surpasses the state-of-the-art update methods on different 3D medical image segmentation datasets including the images of brain tumor, heart and prostate. We summarize our contributions as follows:
- We formulate the interactive image segmentation task as an MDP and propose a novel voxel-wise interactive segmentation framework based on MARL for 3D medical images, enabling more effective utilization of user interaction.
- We propose to reserve the prediction uncertainty via the segmentation probability, which can enrich the information of previous segmentations and lead to a more precise and finer adjustment.
- Extensive experiments show that the segmentation is significantly improved over the iteration sequence with only a few interactions and a rapid convergence, by considering the relative gain between two successive steps.
Related work
============
Interactive image segmentation has been widely applied to both natural [@boykov2001interactive; @xu2016deep] and medical images [@rajchl2016deepcut; @wang2018interactive; @wang2018deepigeos]. “Interactive” refers that the operator provides some hints to the segmentation model to achieve a better result. This section will briefly review the existing works.
Graph-based interactive image segmentation
------------------------------------------
Traditional methods make use of low-level features such as the histogram and similarities between pixels. GraphCut [@boykov2001interactive] and GrabCut [@rother2004grabcut] incorporate user hints into Max-Flow Min-Cut algorithm[@boykov2004experimental]. DenseCRF [@krahenbuhl2011efficient] considers pixel relations from neighbors to all pixel pairs. [@criminisi2008geos] proposes to use geodesic distance to calculate the distance between pixels, which is sensitive to contrast and suitable for medical images. [@wang2016slic] introduces a segmentation method for fetal MRI by learning from user annotations in only one slice.
CNN-based interactive image segmentation
----------------------------------------
Recently, using convolutional neural networks (CNNs) has become popular for computer vision problems. Many CNN-based methods have developed for interactive image segmentation tasks. [@xu2016deep] is the first one to use CNN in interactive image segmentation. [@rajchl2016deepcut] replaces the Gaussian mixture model (GMM) in GrabCut with a CNN for MRI segmentation. Another work, 3D U-Net [@cciccek20163d] learns to produce a complete segmentation from sparsely-annotated slices of one 3D medical image. In order to save the budget of the initial user hints, the following methods, known as update methods, choose to take an automatically-produced segmentation as part of the model initial input. [@wang2018deepigeos] proposes a two-stage method called DeepIGeoS to refine the segmentation using the initial coarse segmentation in input. However, the refined segmentations after the first step cannot be efficiently used in this model. [@bredell2018iterative] extends DeepIGeoS to an iterative version: Inter-CNN, which iteratively refines the previous refined binary prediction in both training and testing stages. One of their problems is the ignorance of the dynamic process for successive interactions. Another problem is the accuracy loss caused by the quantization from probability to binary segmentation.
RL-based interactive image segmentation
---------------------------------------
There are also some methods using RL to explicitly model the dynamic process in interactive image segmentation tasks. SeedNet [@song2018seednet] uses an RL agent to simulate the user behavior which gives the hints to the segmentation model. Since our method uses RL as the segmentation model to predict the segmentation, our method is orthogonal with it. Polygon-RNN [@castrejon2017annotating] identifies the object segmentation as a polygon. Their model produces vertices sequentially until the polygon is closed. The user can contribute by adjusting the vertices. Based on this work, Polygon-RNN++ [@acuna2018efficient] develops a faster and more accurate algorithm by combining RL with graph neural network. However, these polygon-based methods cannot be applied to our tasks because of the incompatibility of the 3D images with polygon segmentation, and the extreme large action space even with the meshing strategy.
Methodology
===========
In this section, we formulate the interactive image segmentation as a MDP and propose a novel MARL-based interactive medical image segmentation method to exploit the interaction information more efficiently.
Overview
--------
In our work, we propose an iteratively-refined framework based on update methods, as shown in Fig. \[overview\], which iteratively refines a coarse initial segmentation by integrating user interactions in order to get a more precise segmentation result. The initial segmentation can be obtained from any accessible segmentation methods.
As discussed in Sections 1 and 2, the main issue of the existing supervised learning-based algorithms is that they split the whole image refinement process into isolated steps. To address this problem, we adopt RL to explicitly capture the relation between successive predictions by designing the reward as the relative improvement. As the large state space and action space of voxel-wise prediction and the necessity of the collaborations between interdependent voxels, we use the idea of MARL: each voxel in a 3D image is regarded as an agent. The work PixelRL [@furuta2019fully] also sees each pixel as an agent, but it focuses on general image processing tasks without human interaction. In contrast, the interactive image segmentation task is more suitable to adopt RL due to its intrinct sequentiality. Unlike dealing with non-interactive image processing tasks, we aim to better consider and effectively utilize external supervision signal from the user during the interaction.
![The flow chart of iteratively-refined interactive image segmentation approach. Given a coarse segmentation, the method iteratively refines it with user interaction until the fine segmentation is good enough.[]{data-label="overview"}](fig/overview.eps){width="3.2in"}
{width="90.00000%"}
Fig. \[framework\] introduces the framework of our proposed method IteR-MRL. By utilizing the original 3D image, the previous segmentation probability and the interactive information as the state, the actor network in the middle gives an update to the segmentation probability and produces a new one. Note that the previous segmentation probability is come from the previous update iteration and the interactive information is the hint map transformed from user hints which will be introduced in Subsection 3.2. The actor network outputs actions of agents which adjust previous segmentation probability and generate current segmentation probability. Afterwards, there are two subsequent operations for the current segmentation probability. On the one side, it gives back a reward signal to the network, by computing the relative gain between previous and current cross entropy based on ground truth and successive segmentation probabilities, for parameter updating. On the other side, it is presented to the user and the user provides feedback, clicks on object or background for wrongly predicted areas. The click is represented as a red point on Fig. \[framework\], which is enlarged for visualization. Generally, with a coarse segmentation probability produced by the initial method (any segmentation method) as its initial segmentation, IteR-MRL iteratively refines the segmentation probability until the segmentation is satisfactory. In addition, the actor network employed here is designed for MARL and it regards voxels on the 3D image as agents who collaborates with each other.
It should be noticed that instead of quantizing the segmentation probability to binary segmentation prediction like previous methods [@bredell2018iterative; @wang2018deepigeos], here we directly use the segmentation probability as the previous segmentation information and feed it into the model. The segmentation probability is introduced to enrich the previous segmentation information and achieve more accurate results. With the segmentation probability, we can derive an adjustment action space leads to a more precise and finer segmentation comparing with the binary segmentation quantization. Specifically, we can adjust the segmentation probability at each step and choose the best adjustment magnitude from a set of various scales. The adjustment action for one agent in MARL model is based on both its own and the neighbors’ states.
Multi-agent RL framework for interactive image segmentation
-----------------------------------------------------------
In this subsection, we describe the MARL setting for interactive image segmentation. Let $\bm{x} = (x_1, \cdots, x_N)$ be one arbitrary image in the dataset and $x_{i}$ is the $i_{th}$ voxel of $\bm{x}$. We treat each $x_{i}$ as an agent whose policy is defined as $\pi_{i} (a_i^{(t)}|s_i^{(t)})$. $s_i^{(t)}$ and $a_i^{(t)}$ are the state (image, previous segmentation probability, user interaction) and action (adjustment to previous probability) for $x_{i}$ at the step $t$; $a_i^{(t)} \in \mathcal{A}$ and $\mathcal{A}$ is the action set; $s_i^{(t)} \in \mathcal{S}$ and $\mathcal{S}$ is the state set. By using convolutional kernels, one agent can access to its neighbors’ states as well, where neighbors are considered as near voxels.
From the point of view of the whole image, the previous segmentation is refined to a new one. By taking the global action $\bm{a}^{(t)} = (a_1^{(t)}, \cdots, a_N^{(t)})$, the image agent transfers to the global state $\bm{s}^{(t+1)} = (s_1^{(t+1)}, \cdots, s_N^{(t+1)})$ and gets the global reward $\bm{r}^{(t)} = (r_1^{(t)}, \cdots, r_N^{(t)})$.
We now define the state, action and reward of a single agent $x_{i}$ in IteR-MRL.\
**State.** For our problem formulation, the state for voxel agent $x_{i}$ at the step $t$ is the concatenation of its voxel value $b_{i}$, its previous segmentation probability $p_{i}^{(t)}$ to be object label and its two values on hint maps $h_{+,i}^{(t)}$ and $ h_{-,i}^{(t)}$: $s_{i}^{(t)}=[b_{i}, p_{i}^{(t)}, h_{+,i}^{(t)}, h_{-,i}^{(t)}]$ with $p_{i}^{(t)} \in [0,1]$. For the initial state $s_{i}^{(0)}$, the initial coarse segmentation probability denotes initial probability $p_{i}^{(0)}$.
Now we discuss the generation of a whole hint map. Concerning the user interaction at step $t$, the hint map $\bm{h}^{(t)}$ is transformed from the user’s hints which are in the form of click points. By giving a hint point through a single click, the user indicates that the area around it is one error region. Intuitively, the closer one point is to the hint point, the more likely its label is mispredicted. Hence, the hint map is introduced to show the radiation area of the hint and spread the local interaction to the whole image. The number and positions of hints are chosen according to the user interaction rule. Actually, there are two channels of hint map both with the same size as the image: the object hint map $\bm{h}_{+}^{(t)}$ and the background hint map $\bm{h}_{-}^{(t)}$, respectively generated from the object hint set $hs_{+}^{(t)}$ (hints on object) and the background hint set $hs_{-}^{(t)}$ (hints on background). Hence, the user hint map is the concatenation of these two hint maps: $\bm{h}^{(t)}=[\bm{h}_{+}^{(t)}, \bm{h}_{-}^{(t)}]$. For one hint map $\bm{h}_{\ell}^{(t)}, \ \ell \in \{+, -\}$, we define that $\bm{h}_{\ell}^{(t)} = (h_{\ell,1}^{(t)}, \cdots, h_{\ell,N}^{(t)})$. The element $h_{\ell,i}^{(t)}$ on the hint map $\bm{h}_{\ell}^{(t)}$ is calculated by the minimum distance between $x_i$ and the corresponding hint set $hs_{\ell}^{(t)}$: $$% \label{min distance}
h_{\ell,i}^{(t)} \ = \ \min_{\forall x_j \in hs_{\ell}^{(t)}} \mathcal{M}(x_i, x_j),\\\ \ell \in \{+, -\},$$ where $\mathcal{M}$ is a function to measure the distance between two voxels. Previous related works adopt several distance-measuring methods including geodesic [@criminisi2008geos], Gaussian and Euclidean distance. In the paper, we use the geodesic distance-based hint map to measure distances. The distance between two voxels is the minimum value of the sum of color gradients across all the paths connecting these two voxels. (See the hint map in Fig. \[framework\].)\
**Action.** While previous works [@bredell2018iterative; @wang2018deepigeos] output directly the segmentation probability from the network, we here predict the adjustment amount based on the previous probability as actions to make the result more stable without abrupt changes. The action $a_{i}^{(t)} \in \mathcal{A}$ for $x_{i}$ at time step $t$ is to adjust the previous segmentation probability $p_{i}^{(t)}$ by a certain amount $a_{i}^{(t)}$. Hence, the segmentation probability $p_{i}^{(t+1)}$ after taking the action $a_{i}^{(t)}$ is: $$\begin{aligned}
% \label{segmentation proba}
p_{i}^{(t+1)} \ &=\ \mathcal{C}_0^1(p_{i}^{(t)} + a_{i}^{(t)}),\\
\label{clip}
\mathcal{C}_a^b(x) \ &=\ \min(\max(x, a), b),\end{aligned}$$ where $\mathcal{C}_a^b(x)$ is a function to clip the value of $x$ from $a$ to $b$. $p_{i}^{(t+1)}$ is constrained to $[0,1]$ for it represents a probability. The action set $\mathcal{A} = \{\mathcal{A}_k\} \ (k = 1,2, \cdots, K)$ contains $K$ actions, allowing the agent to adjust the probability to various degrees under different situations. For example, it is reasonable to make a larger adjustment to a voxel when it is closer to a hint click. Additionally, one voxel tends to take one certain adjustment action when most of its neighbor voxels choose this action.\
**Reward.** To improve the efficiency of exploration, we design a relative cross-entropy gain-based reward to update the model in a constrained direction. Specifically, the reward is designed as the relative improvement from the previous segmentation to the current one, which is the decreased amount of the cross entropy $\mathcal{X}_{i}$ between the ground truth $y_{i}$ and the segmentation probability $p_{i}$: $$\label{reward}
r_{i}^{(t)} \ =\ \mathcal{X}_{i}^{(t-1)} - \mathcal{X}_{i}^{(t)},$$ where $$\label{cross entropy}
\mathcal{X}_{i}^{(t)} \ =\ -y_{i}\log(p_{i}^{(t)})-(1-y_{i})\log(1-p_{i}^{(t)}).$$
With , the agent gets a positive reward in the case its probability moves closer towards the true voxel label and vice versa. Instead of a distant goal, the relative gain provides the agent with a baseline to compare and surpass.
In general, the accumulated reward of one interactive sequence is $$% \label{sequence reward}
R_i \ = \ \textstyle\sum_{t=1}^{T}\gamma^{t-1} r_{i}^{(t)},$$ where $T$ is the total step number and the discount factor $\gamma$ takes a value in $(0, 1]$.
Network and training
--------------------
For fair comparison, the interactive network architecture of [@wang2018deepigeos] named R-net is adopted as the backbone to our algorithm and all other baseline methods. We adapt the network to the one in Fig. \[architecture\] in order to fit the RL training algorithm: asynchronous advantage actor-critic (A3C) [@mnih2016asynchronous]. The network firstly uses three 3D convolutional blocks to extract low-level features. Then, the network is divided into two heads: policy head and value head. Both of the heads have three 3D convolutional blocks to extract specific high-level features. The functionality of the policy head is to predict the distribution of action probabilities under a known state. In our case, given the image, hint maps and previous segmentation probability, the policy head predicts how likely it is to take each scale of adjustment to the previous segmentation probability. The functionality of the value head is to estimate the value of the current state. Specifically, the value head evaluates how good the current combination of the image, hint maps and the previous segmentation probability is.
We respectively use $\theta_p$ and $\theta_v$ to denote the parameters of the policy and value heads. The input of the network is the state at time step t: $\bm{s}^{(t)}$. The value head outputs the estimated value of the current state $V(\bm{s}^{(t)})$. The gradient for $\theta_v$ is computed by: $$\begin{aligned}
% \label{gradient v}
d\theta_v \ &= \ \nabla_{\theta_v}A(\bm{s}^{(t)}, \bm{a}^{(t)})^2,\\
% \label{advantage}
A(\bm{s}^{(t)}, \bm{a}^{(t)}) \ &= \ \sum_{k=t}^{T}\gamma^{k-t} \overline{r}^{(k)} - V(\bm{s}^{(t)}),
% \label{r}
% R^{(t)} \ &=\ \textstyle\sum_{k=t}^{\infty}\gamma^{k-t} \overline{r}^{(k)},\end{aligned}$$ where $\overline{r}^{(k)}$ is the mean reward of all voxels at time step $k$. $A(\bm{s}^{(t)}, \bm{a}^{(t)})$ is the advantage at time step $t$ of taking $\bm{a}^{(t)}$ in condition of state $\bm{s}^{(t)}$, which indicates the actual accumulated reward without being affected by the state and reduces the variance of gradient. The policy head outputs the action policy $\pi(\bm{a}^{(t)}|\bm{s}^{(t)})$, which is the probabilities of taking each action $\bm{a}^{(t)}$. The gradient for $\theta_p$ is computed by: $$% \label{gradient p}
d\theta_p \ = \ -\nabla_{\theta_p}\pi(\bm{a}^{(t)}|\bm{s}^{(t)})A(\bm{a}^{(t)}, \bm{s}^{(t)}).$$ The two heads are jointly trained in an end-to-end manner.
![The network architecture for IteR-MRL. The policy and value heads share the low-level features and extract their own high-level features.[]{data-label="architecture"}](fig/architecture.eps){width="3in"}
Experiments
===========
Datasets
--------
In our paper, we do experiments on three 3D MRI datasets. Each image is cropped based on its non-zero region before used. For each dataset, we access all the image cases having ground truth and split them into several sets. The initial method is defined as the segmentation method of producing the initial coarse segmentations. If we use the images trained on initial method again in update method, the coarse segmentation probability (initial segmentation probability for update method) will be too perfect to be refined in update method because these images have already seen the ground truth in initial method. Therefore, we propose a new way for the dataset splitting: the dataset is split into three parts, two training sets with an equal amount of images and one testing set. In detail, we randomly selected $N_{train}$ cases as the training set for initial method forming $D_{train1}$ and randomly selected another $N_{train}$ cases in the remaining dataset as the training set for update method forming $D_{train2}$. The remaining $N_{test}$ cases are used as testing forming $D_{test}$. Note that the initial segmentation probabilities data for $D_{train2}$ in update method are obtained by testing $D_{train2}$ with initial method. The three datasets are as follows:\
**BraTS2015.** Brain Tumor Segmentation Challenge 2015 (BraTS) [@menze:hal-00935640] provides a dataset for brain tumor segmentation in magnetic resonance images. We use Fluid-attenuated Inversion Recovery (FLAIR) images which contain 274 cases and only segment the whole brain tumor. We set $N_{train}$ as 117 and $N_{test}$ as 40.\
**MM-WHS.** Multi-Modality Whole Heart Segmentation (MM-WHS) [@zhuang2016multi] contains multi-modality whole heart images covering the whole heart substructures. We use the 20 MRI cases and segment the left atrium blood cavity. We set $N_{train}$ as 8 and $N_{test}$ as 4.\
**NCI-ISBI2013.** NCI-ISBI 2013 Challenge [@bloch2015nci] aims at automated segmentation of prostate structures. It provides 80 prostate gland MRI data. We set $N_{train}$ as 32 and $N_{test}$ as 16.
Settings
--------
**Evaluation metrics.** Normally, medical image segmentation is evaluated by the dice score: $$% \label{dice}
Dice(S_p, S_g) \ = \ \frac{2|S_p \cap S_g|}{|S_p| + |S_g|},$$ where $S_p$ represents the predicted segmentation and $S_g$ represents the ground truth. $|\cdot|$ is the number of voxels in the area.
As we study the interactive image segmentation task, we consider not only the dice score but also the user click number. Our goal is to get a high dice score with a small number of user clicks.\
**User simulation.** Since it would require large human resources to conduct the experiments with real physicians, we simulate user clicks like other works. While previous works usually give many clicks ($\approx 40$) for training but a few clicks for testing, our interaction policies for training and testing are consistent. Hence, the training setting is similar to that of testing in order to reduce the bias between training and testing. In one training/testing sequence of an image, we give $N_{click}$ clicks each step. Specifically, the clicks are selected as the centers of the largest $N_{click}$ error regions. In addition, a small disturbance $\epsilon_{noise}$ is added to each click point to force the model to be robust and also make it imitate the behavior of a real user.\
**Implementation details.** For the preprocessing, all the images are normalized by the mean and the standard variation of the whole dataset $D = [D_{train1}, D_{train2}, D_{test}]$. Each image is cropped by the bounding box based on its non-zero region with an extension of $[0, 10]$ voxels and then resized to the size of $55 \times 55 \times 30$. Data augmentation involves flipping in three directions and random rotation with angle range $[-\pi/8, \pi/8]$ in three directions.
As the proposed IteR-MRL can be easily adapted to the interaction-free mode, we firstly train a pure segmentation model IteR-MRL0 as the pretrained model for IteR-MRL. IteR-MRL0 is trained for 1000 epochs and IteR-MRL fine-tunes on IteR-MRL0 for 500 epochs. The learning rate adopts the step decay schedule with an initial learning rate $10^{-4}$. Parameter setting is as follows: $T=5$, $N_{click}=5$, $\gamma=0.95$, $\epsilon_{noise}=[-3,3]^3$. We use Adam algorithm for optimization with minibatch size 1.
The model training time with one Nvidia Titan X GPU varies from several hours to two days for different datasets. The average inference time for each update step is 894ms, which includes 424ms of the interaction simulation time.
Results
-------
For fair comparison, we apply denseCRF to all the models compatible with CRF as the final refinement processing.\
**Comparisons with state-of-the-art methods.** We compare IteR-MRL with three state-of-the-art methods: Min-Cut [@krahenbuhl2011efficient], DeepIGeoS(R-Net) [@wang2018deepigeos] and InterCNN [@bredell2018iterative].
\[tab:initial\]
![Performance improvement shown by curves[]{data-label="plot"}](fig/plot.eps){width="3.3in"}
In Table \[tab:initial\], the update methods receive the coarse segmentations from four different initial segmentation methods: BG (set all voxel labels to background), V-Net [@milletari2016v], HighRes3DNet [@li2017compactness] and DeepIGeoS(P-Net) [@wang2018deepigeos]. The experimental results show that IteR-MRL achieves better performances than the other three methods under each initial method, which shows the robustness and generalization of our method.
![Visualization of different update methods[]{data-label="update-compare"}](fig/ud-compare.eps){width="2.8in"}
{width="90.00000%"}
To validate whether considering the relative gain between successive predictions can result in rapid improvements, we also analyze the performance improvement during one refinement sequence in Table \[tab:iteration\]. We use V-Net here as the initial method (77.15). For the first refinement step, all the update methods have significant improvements in performance (from $+2.37$ to $+8.47$). Starting from the second step, most performances have encountered stagnation (very little improvement) though with newly-added user hints. DeepIGeoS(R-Net) even has a degradation ($-0.01$) at the third step. While the other methods improve slowly at each refinement step, IteR-MRL has a relative high improvement, which proves the effectiveness of considering the relation gain between successive predictions. The large improvement at each refinement step also leads to a good result ($88.53$) in the end. In addition, we notice that IteR-MRL’s performance at the second step has already surpassed the others’ final performances, achieving a reduction of user click number. Fig. \[plot\] provides a global view of performance improvement in one interactive sequence.
Fig. \[update-compare\] gives the visualization of different update methods using V-Net as the initial method. Specifically, we visualize the refined segmentations after five refinement steps. It can be observed that while the other methods tend to produce a rather smooth boundary, IteR-MRL performs better in capturing edge details.
\[tab:data\]
The above results are obtained from the experiments with the dataset BraTS2015. More experiments are also conducted on the other two datasets MM-WHS and NCI-ISBI2013 in Table \[tab:data\] to verify the robustness, with the initial method V-Net. The results prove that IteR-MRL has stable performances on various types of datasets.\
**Ablation study.** We analyze the effect of different action sets to the algorithm performance in Table \[tab:settings\]. Specially, when the action set only contains ${\pm1.0}$ (line 1), the segmentation probability becomes binary, because the segmentation probability can only take the values 0 and 1. The rest action sets are all designed for the states containing segmentation probability. The influence of the action value and the action number are both analyzed. For the influence of value, we fix the number of actions and let action values vary: we try $\pm1.0$, $\pm0.4$, $\pm0.2$ and $\pm0.1$ (line 1, 2, 3, 4). Comparing the states adopting segmentation probability to those of binary prediction, it can be found that binary prediction has a poor performance caused by the loss of prediction uncertainty. In addition, the results show that small action values have better performances than the larger ones. The reason is that a small action value allows IteR-MRL to make more detailed adjustments, but a large adjustment may over-behave and never reach some specific states. For the influence of the action number, we gradually add new actions to the action set (line 4, 5, 6, 7). It can be observed that abundant actions lead to a better performance by providing IteR-MRL with various degrees of adjustment. In the case with a high confidence, IteR-MRL tends to take a large adjustment, which speeds up the refinement convergence. However, the addition of $\pm1.0$ relatively damages the performance because an adjustment of $\pm1.0$ is too extreme for most of the cases. In general, we learn that the combination of small and large actions except for $\pm1.0$ works best and finally choose $\mathcal{A}=\{\pm0.1, \pm0.2, \pm0.4\}$ as the ideal action set in our model. Note that we have also thought about a continuous action space, but the experimental result shows that it is difficult to train and converge in a continuous action space. In addition, since the final prediction presented to the user is 0 or 1 for each voxel, we are not concerned about the inability to reach the optimal final state with a discrete action space.
As we know, the interaction and the model can both lead to the improvement of performance. We now analyze how much the interaction contributes to the performance by changing the interaction strategy. In addition to the good interaction used before, two more comparative experiments are done in Table \[tab:interaction\]. The one without interaction is to always fill the hint maps with random noise and the model will not receive any new interactive information. The other one with bad interaction is to randomly choose the user click points among all the voxels. In this case, the interaction may pass the wrong message to the model. As a result, we find that the meaningful interaction does help greatly improve the performance. It can also be observed that the one without interactions still has some gain of performance, which may come from the iterative model itself. Moreover, the degradation of the one with bad interactions indicates that ineffective interaction can damage the performance.
Fig. \[ud-ablation\] presents the visualization for the relation between predictions and hints. Fig. \[ud-ablation\](a) shows the influence of user interaction on prediction and hint maps. Since the data is 3D, we show the slice with click (the middle row) and its two neighbor slices (rows on both sides). The red parts on hint maps are the recommended object regions. We find that the proposed algorithm can successfully correct the local region around the user click (the red point). Besides, the corresponding regions on neighbor slices are also improved. In Fig. \[ud-ablation\](b), we observe the change of predictions and hint maps in one interactive sequence which contains five steps. The user clicks are not shown because the slices with clicks vary at each step and we only focus on changes of one specific slice. With the indications of hint maps, IteR-MRL succeeds in refining the initial prediction step by step.
Conclusion
==========
In this paper, we propose a novel iteratively-refined interactive segmentation method for 3D medical images using multi-agent reinforcement learning. The method explicitly models the dynamic process of interactive image segmentation task in order to get a rapid segmentation improvement at each iteration. In addition, it augments the state space of MDP with the segmentation probability, allowing the agent to make finer adjustments than the conventional binary approach. The experimental results show that it performs better than the state-of-the-art methods and it is robust to different initial segmentations and various datasets.
[^1]: Equal contribution
| null |
minipile
|
NaturalLanguage
|
mit
| null |
One of the most significant developments in semiconductor technology in recent years has been the increased use and importance of compound semiconductors. Particularly significant are the Group III-V compound semiconductor devices composed of elements of Groups III and V of the periodic table, such as gallium arsenide (GaAs) and indium phsophide (InP). Compound semiconductors are used in such devices as lasers, light-emitting diodes, microwave oscillators and amplifiers, and various types of light detectors.
Most such commercial use of compound semiconductors requires the growth of large single-crystal ingots from which wafers can be cut for the subsequent fabrication of useful devices. One of the more promising methods for such crystal growth is the Vertical Gradient Freeze (VGF) method, particularly the VGF method defined in the U.S. pat. No. of W. A Gault, 4,404,172, granted Sept. 13, 1983 and assigned to Western Electric Company, Inc., which is hereby incorporated herein by reference. According to this method, polycrystalline starting material is placed in a vertically extending crucible including a small cylindrical seed well portion at its bottom end which snugly contains a monocrystalline seed crystal. Initially, the starting material and a portion of the seed crystal are melted. The power to the system is then reduced in such a manner that freezing proceeds vertically upwardly from the seed crystal. One major advantage of the VGF method is that, by using low thermal gradients, crystals with very low dislocation densities can be produced. Another advantage is that the crystal growth rate be changed with no effect on the crystal diameter.
It is well-known that the Group III-V compounds tend to dissociate at higher temperatures, with the more volatile Group V element escaping into the vapor phase. Several approaches have been developed to prevent or retard this tendency during crsytal growth. In one approach, escape of the more volatile Group V component is retarded by providing a vapor pressure of Group V vapor over the melt from a sperately heated reservoir of Group V material within the sealed growth container. It is also knwon that Group V material loss from the melt may be retarded with the use of any of various materials such as boric oxide (B.sub.2 O.sub.3), barium chloride (BaCl.sub.2), or calcium chloride (CaCl.sub.2) which act as diffusion barriers. Such additives, having a lower density than the molten indium phosphide, rise to the surface, encapsulate the melt, and, together with an inert gas pressure in the vessel, can contain the volatile vapors; see, for example, the paper of "Growth of Single Crystals of GaAs in Bulk and Thin Film Form," by B. A. Joyce, included in the book, Crystal Growth, edited by B. R. Pamplin, Pergamon Press, 1975, pp. 157-184 at p. 165.
The use of boric oxide as a diffusion barrier can sometimes lead to a fully stoichiometric crystal. But in most cases there is some loss of the volatile Group V material, resulting in excess Group III inclusions in the last-to-freeze portion of the crystal. "Inclusion," as is known in the art, is a small volume within the crystal structure having an excess of one of the constituents or of impurity atoms, e.g., an excess of indium in indium phosphide.
The use of a separately heated reservoir of Group V material has been found to require relatively large amounts of the Group V material for each crystal growth run. The pressure vessel must be cleaned frequently because most of the Group V material sublimes and deposits on the cold surface of the pressure vessel. Also, the requirement of a second heater consitutes an added cost and complication to the system.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
---
abstract: 'We calculate the integral cohomology ring structure for various members of the Bianchi group family. The main tools we use are the Bockstein spectral sequence and a long exact sequence derived from Bass-Serre theory.'
address: |
Lafayette College\
Department of Mathematics\
Easton, PA 18042
author:
- Ethan Berkove
title: The Integral Cohomology of the Bianchi Groups
---
Introduction
============
In the calculation of $H^*(G;M)$, the cohomology of a group $G$ with coefficients in $M$, it is standard practice to let $M$ be a field in order to simplify the mathematics. The integral cohomology ring, with $M = {\mathbb{Z}}$, can be much harder to determine, particularly when the group contains a lot of torsion. However, it many cases it is still possible to get a complete answer. In this paper we calculate the integral cohomology rings for a number of Bianchi groups. These groups are particularly amenable to calculation as their torsion occurs only at the primes 2 and 3, their mod-2 cohomology rings and integral homology groups are known, and their 3-primary cohomology is easily calculable. The groups themselves can be built in stages out of finite groups using amalgamated products and HNN extensions. Therefore, their group cohomology is closely related to the cohomology of these finite pieces. This paper completes a calculation that has partially appeared in a number of places: For various members of the Bianchi group family, the integral homology (additive structure) can be found in [@S-V], the rational homology in [@V], and the mod-2 cohomology in [@Be].
The main tools we use at the prime 2 are the Bockstein spectral sequence and long exact sequences that come from Bass-Serre theory. At the prime 3 we perform the calculations solely using long exact sequences. This paper is organized as follows. In the next section we describe the Bianchi groups and the techniques we will need to determine their integral cohomology rings. In the third section we apply these techniques to determine some intermediate results. In the fourth section we give complete details for the calculation of $H^*(G;{\mathbb{Z}})$ where $G = {{\Gamma}_{2}}, {{\Gamma}_{6}}$. In the final section we do the cases $G = {{\Gamma}_{1}}, {{\Gamma}_{3}}, {{\Gamma}_{5}}, {{\Gamma}_{7}}, {{\Gamma}_{10}}, {{\Gamma}_{11}}$.
Background
==========
For $d$ a square-free integer, let ${{\mathcal O}}_d$ denote the ring of algebraic integers in the imaginary quadratic extension field of the rationals, ${\mathbb Q}(\sqrt{-d})$. As a ${\mathbb{Z}}$-module, ${{\mathcal O}}_d \cong {\mathbb{Z}}\oplus {\mathbb{Z}}\ \omega$, where $\omega = \sqrt{-d}$ when $d \equiv 3$ mod $4$ and $\omega = \frac{1 + \sqrt{-d}}{2}$ when $d \equiv 1,2$ mod $4$. The Bianchi groups $\Gamma_d$ are the projective special linear groups $PSL_2({{\mathcal O}}_d)$, and can be thought of as the natural generalizations of the modular group, ${PSL_2({\mathbb Z})}\cong {\mathbb{Z}}/2 \ast {\mathbb{Z}}/3$.
The Bianchi groups are excellent candidates for integral cohomology calculations. Torsion in Bianchi groups occurs at the primes 2 and 3, so one only needs to perform two $p$-primary calculations. Furthermore, the most complicated torsion occurs at the prime $p=2$, and mod-2 cohomology rings (as modules over the Steenrod algebra) have already been calculated for the Bianchi groups $d = 1, 2, 3, 5, 6, 7, 10$ and $11$ in [@Be].
Starting with mod-$p$ cohomology information, the $p$-primary component of the integral cohomology can be reconstructed using a handy tool: the Bockstein spectral sequence. This spectral sequence, which first appeared in [@Browd], is associated to the following exact couple. $$\ldots {\buildrel \delta \over \longrightarrow} H^*(X;{\mathbb{Z}}) {\buildrel {(\times p^*)} \over \longrightarrow} H^*(X;{\mathbb{Z}}) {\buildrel red_{p^*} \over \longrightarrow} H^*(X;{\mathbb F}_p) {\buildrel \delta \over \longrightarrow} H^{*+1}(X;{\mathbb{Z}}) {\buildrel {(\times p^*)} \over \longrightarrow}\ldots$$
The first differential is $d_1 = red_{p^*} \circ \delta$. In fact, $d_1 = \beta$, the classical Bockstein homomorphism, which can be identified with the Steenrod square $Sq^1$ when $p=2$. Therefore, if $Sq^1 (w) \neq 0$ for some class $w \in H^n(X;{{\mathbb F}_2})$, then $\delta (w)$ represents some cohomology class $\hat{w} \in H^{*+1}(X;{\mathbb{Z}})$ of order 2. Elements that appear in the integral cohomology at different pages of the spectral sequence have different orders; the $r^{th}$ differential $d_r$ can be identified with the $r^{th}$ order Bockstein homomorphism [@McC] $$\beta_r :H^*(X; {\mathbb{Z}}/2^r {\mathbb{Z}}) {\buildrel \times 2^{r-1} \over \longrightarrow} H^{*+1}(X; {\mathbb{Z}}/2^r {\mathbb{Z}}).$$ Thus, elements that survive to the $E_r$ page have order at least $2^r$. The Bockstein spectral sequence converges to $(H^*(X;{\mathbb{Z}}) / \text{$p$-torsion} ) \otimes_{{\mathbb{Z}}} {\mathbb F}_p$. We use this sequence to calculate group cohomology via the identification $H^*(G;M) = H^*(BG;M)$, where $BG$ is the classifying space of $G$.
For the Bianchi groups we consider, the Bockstein spectral sequence applied to the calculations in [@Be] almost collapses at the $E_1$ page, with at most one torsion class surviving to the $E_2$ page. Thus, almost all 2-torsion in the resulting integral cohomology rings has order 2, a result consistent with the calculations in [@S-V]. Although it would be convenient if this were the only tool needed in this paper, the Bockstein spectral sequence is less effective at identifying torsion-free classes, where knowledge of all Bockstein homomorphisms is required. Further, we need a technique to determine the 3-primary cohomology ring.
Fortunately, there is another technique. The Bianchi groups we consider in this article are built in stages using amalgamated products and HNN extensions from copies of ${\mathbb{Z}}$ and five finite subgroups: the cyclic groups of order two and three (${\mathbb{Z}}/2$ and ${\mathbb{Z}}/3$), the Klein four group (${{\bf D}_2}\cong {\mathbb{Z}}/2 \times {\mathbb{Z}}/2$), the symmetric group on three letters (${{\bf S}_3}$), and the alternating group on four letters (${{\bf A}_4}$) [@G-S]. We show how the cohomology of a group built via these constructions depends on the cohomology of its component subgroups. This then allows us to determine the 3-primary and rational cohomology rings. At the prime 2, this approach is also helpful for identifying the few classes of order 4 in the 2-primary cohomology ring. We summarize some useful results, starting with the definition of an HNN extension. A more thorough treatment can be found in [@Be].
Let $G_1$ be a group, $H$ be a subgroup and $\theta:H \rightarrow G_1$ be a monomorphism. Then an [*HNN extension of $G_1$*]{} is a group $$\Gamma = \{ t,G_1 \mid t^{-1}at = \theta(a), a \in H \}.$$ $\Gamma$ is denoted by $G_1 \ast_H$. $G_1$ is called the [*base*]{} and $H$ is called the [*associated subgroup*]{}.
When $\Gamma \cong G_1 \ast_H G_2$ is an amalgamated product or $\Gamma \cong G_1 \ast_H$ is an HNN extension, Bass-Serre theory [@S] provides a long exact sequence. $$\cdots {\buildrel \alpha \over \longrightarrow}H^{j-1}(H)
{\buildrel \delta \over \longrightarrow}H^j(\Gamma) {\buildrel \rho \over \longrightarrow}\oplus H^j(G_i)
{\buildrel \alpha \over \longrightarrow}H^j(H) {\buildrel \delta \over \longrightarrow}\cdots
\label{LES}$$ The maps above can be explicitly described. The map $\rho$ is a sum of restriction maps to each summand. When $\Gamma$ is an amalgamated product, $\alpha$ is the difference of restriction maps, $\alpha = res^{G_1}_H - res^{G_2}_H$. When $G$ is an HNN extension, the “twisting” induced by $\theta$ comes into play, and $\alpha = res^{G_1}_{H} - \theta^*$.
Whenever $\Gamma$ an HNN extension, $\rho$ is an isomorphism in degree zero. Consequently, $\alpha$ factors through zero and $\delta$ is an isomorphism, generating a torsion-free class in degree 1. We call such classes “HNN classes.” By degree considerations, they are exterior.
The map $\rho$ respects cup products, since it is a restriction map in the case of an HNN extension and a sum of restrictions in the case of an amalgamated product. Thus, by exactness of $H^*(\Gamma) {\buildrel \rho \over \longrightarrow}\oplus H^*(G_i) {\buildrel \alpha \over \longrightarrow}H^*(H)$, we have that $im(\rho) \cong ker(\alpha)$, and this is an [**isomorphism of rings**]{}. In most of our calculations the map $\alpha$ is surjective, so the long exact sequence of Formula \[LES\] breaks into short exact sequences. Hence the ring isomorphism determines most of the cup product structure of $H^*(\Gamma)$, which can be used to determine cup products in the cases where the Bockstein spectral sequence information is incomplete or ambiguous.
In those cases where $\alpha$ is not surjective, classes can also arise in $H^*(\Gamma)$ from the image of $\delta$. The following lemmas describe how to determine products with these classes.
[@Sp]. For $u \in H^p(\Gamma)$ and $v \in H^q(H),\ \delta (res^{\Gamma}_H u \cup v) = u \cup \delta (v),$ and $\delta (v \cup res^{\Gamma}_H u) = \delta (v) \cup u$. \[compat\]
Classes that arise from the image of $\delta$ have trivial products with each other. \[PY\]
[[*Proof.*]{} ]{}See Theorem 4.8 in [@P-Y] [$\square$]{}
Finally, on occasion we will need to determine restriction maps on integral cohomology. If the restriction map is known at the level of mod-2 cohomology, one can extend the map by appealing to the following long exact sequences of commuting squares.
(10,2.5) (2,2)[$\ldots \longrightarrow H^*(G) {\buildrel {\times p} \over \longrightarrow} H^*(G) {\buildrel red_{p^*}
\over \longrightarrow} H^*(G;{\mathbb F}_p) {\buildrel \delta \over \longrightarrow} \ldots$]{} (2,0)[$\ldots \longrightarrow H^*(H) {\buildrel {\times p} \over \longrightarrow} H^*(H) {\buildrel red_{p^*}
\over \longrightarrow} H^*(H;{\mathbb F}_p) {\buildrel \delta \over \longrightarrow} \ldots$]{} (4,1.7)[(0,-1)[1.2]{}]{} (6,1.7)[(0,-1)[1.2]{}]{} (8,1.7)[(0,-1)[1.2]{}]{} (4.2,1)[$res^G_H$]{} (6.2,1)[$res^G_H$]{} (8.2,1)[$res^G_H$]{}
Initial Calculations
====================
We determine the integral cohomology of the five finite groups that form the basis of the Bianchi group’s cohomology, then apply techniques from the last section to determine some intermediate results. Subscripts on classes denote the degree of the class.
$H^*({\mathbb{Z}}/2;{\mathbb{Z}}) \cong {\mathbb{Z}}[x_2] / (2x_2 = 0)$ and $H^*({\mathbb{Z}}/3;{\mathbb{Z}}) \cong {\mathbb{Z}}[x_2] / (3x_2 = 0)$.
The 2-primary component $H^*({{\bf S}_3};{\mathbb{Z}})_{(2)} \cong H^*({\mathbb{Z}}/2;{\mathbb{Z}}) \cong {\mathbb{Z}}[x_2] / (2x_2 = 0)$, and the isomorphism is induced by restriction. The 3-primary component $H^*({{\bf S}_3};{\mathbb{Z}})_{(3)} \cong {\mathbb{Z}}[x_4] / (3x_4 = 0)$, and this is the image of $res^{{{\bf S}_3}}_{{\mathbb{Z}}/3}$. \[St\]
[[*Proof.*]{} ]{}Consider the short exact sequence $$0 \longrightarrow {\mathbb{Z}}/3 \longrightarrow {{\bf S}_3}\longrightarrow {\mathbb{Z}}/2 \longrightarrow 0.$$ There is a spectral sequence with $E_2^{p,q} \cong H^p({\mathbb{Z}}/2; H^q({\mathbb{Z}}/3;{\mathbb{Z}}))$ converging to $H^*({{\bf S}_3};{\mathbb{Z}})$. The horizontal edge of the spectral sequence is isomorphic to $H^*({\mathbb{Z}}/2;{\mathbb{Z}})$. The vertical edge, however, is the invariants $H^*({\mathbb{Z}}/3;{\mathbb{Z}})^{{\mathbb{Z}}/2}$. The horizontal edge yields the 2-primary part of the cohomology. On the vertical edge, the action is free in dimensions $2 + 4n$ and trivial otherwise [@Bro]. [$\square$]{}
We use the Bockstein spectral sequence to determine the integral cohomology of ${{\bf D}_2}$ and ${{\bf A}_4}$.
$H^*({{\bf D}_2}; {\mathbb{Z}}) \cong {\mathbb{Z}}[y_2, z_2, y_3]$ with relation $y_3^2 = y_2^2 z_2 + y_2 z_2^2$. All classes have order 2.
[[*Proof.*]{} ]{}Let $x_1$ and $y_1$ generate $H^1({{\bf D}_2}; {{\mathbb F}_2})$. Since $Sq^1(x_1) = x_1^2$ and $Sq^1(y_1) = y_1^2$, there are classes $y_2$ and $z_2 \in H^2({{\bf D}_2}; {\mathbb{Z}})$ with $red_{2^*}(y_2) = x_1^2$ and $red_{2^*}(z_2) = y_1^2$. We will identify $y_2$ and $z_2$ with their images under $red_{2^*}$. This directly implies that $y_2$ and $z_2$ generate a rank-2 polynomial algebra over ${\mathbb{Z}}$, as there are no relations between $x_1^2$ and $y_1^2$. There is another non-trivial square, $Sq^1(x_1 y_1) = x_1^2 y_1 + x_1 y_1^2$, which represents the generator $y_3 \in H^3({{\bf D}_2}; {\mathbb{Z}})$. Since $0 \neq (x_1^2 y_1 + x_1 y_1^2)^n \in H^*({{\bf D}_2}; {{\mathbb F}_2})$, $y_3$ is also a polynomial class. Furthermore, as $$(x_1^2 y_1 + x_1 y_1^2)^2 = x_1^2 y_1^2 (x_1^2 + y_1^2)$$ we get the relation $y_3^2 = y_2 z_2 (y_2 + z_2)$ in the integral cohomology. To finish the argument, note that all classes whose square are 0 are also in the image of $Sq^1$. Thus, the spectral sequence collapses at the $E_1$ page, and all classes in the integral cohomology have order 2. A Poincaré series confirms that we have found all the relations. [$\square$]{}
$H^*({{\bf A}_4}; {\mathbb{Z}})_{(2)} \cong {\mathbb{Z}}[y_3,y_4,y_6]$ with relation $y_3^4 + y_4^3 + y_3^2 y_6 + y_6^2 = 0$. All classes have order 2. $H^*({{\bf A}_4}; {\mathbb{Z}})_{(3)} \cong H^*({\mathbb{Z}}/3 ;{\mathbb{Z}})$, and this isomorphism is given by $res^{{{\bf A}_4}}_{{\mathbb{Z}}/3}$. \[Af\]
[[*Proof.*]{} ]{}At the prime 2, start with $H^*({{\bf A}_4};{{\mathbb F}_2}) \cong {{\mathbb F}_2}[u_2,v_3,w_3]$ with relation $u_2^3 + v_3^2 + w_3^2 + v_3w_3=0$ (see [@A-M] for a derivation). The Bocksteins are $Sq^1(u_2) = v_3$, $Sq^1(v_3) = 0$, and $Sq^1(w_3) = u_2^2$. We identify three classes in $H^*({{\bf A}_4};{\mathbb{Z}})$ in low degrees: $y_3$, whose image is $v_3 = Sq^1(u_2)$; $y_4$, whose image is $u_2^2=Sq^1(w_3)$; and $y_6$, whose image is $u_2^3 + v_3w_3 = Sq^1(u_2w_3)$. $Sq^1(u_2v_3) = v_3^2$ is also a non-trivial square, but this class is the image of $y_3^2$ and has thus already been accounted for.
One easily confirms that the classes $v_3$, $u_2^2$, $v_3^2$ and $u_2^3 + v_3w_3$ generate the kernel of $Sq^1$ up to degree 6; therefore, $y_3$, $y_4$, and $y_6$ are of order 2. A Poincaré series argument derived from the Universal Coefficient Theorem implies that these classes satisfy a single relation in degree 12. A straightforward search yields the one above.
At the prime 3, we note that ${\mathbb{Z}}/3$ is self-normalizing in ${{\bf A}_4}$. $H^*({{\bf A}_4}; {\mathbb{Z}})_{(3)}$ maps isomorphically onto the invariants $H^*({\mathbb{Z}}/3; {\mathbb{Z}})^{{{\bf A}_4}}$ ([@Bro], III.10.3), which by direct calculation is $H^*({\mathbb{Z}}/3; {\mathbb{Z}})$. The invariants are also the image of the restriction map. [$\square$]{}
Most of the restriction maps we will need are straightforward. However, $res^{{{\bf A}_4}}_{{\mathbb{Z}}/2}$ is somewhat more complicated.
The map $res^{{{\bf A}_4}}_{{\mathbb{Z}}/2}$ on integral cohomology sends $y_4^i y_6^j$ to $x_2^{2i+3j}$ and all other classes to $0$, where $x_2$ generates $H^*({\mathbb{Z}}/2;{\mathbb{Z}})$. \[AftoZ2\]
[[*Proof.*]{} ]{}We use the commutative square:
(8,2.6) (3.5,2)[$ H^*({{\bf A}_4};{{\mathbb F}_2}) {\buildrel \delta \over \longrightarrow} H^{*+1}({{\bf A}_4}; {\mathbb{Z}})$]{} (3.5,0)[$ H^*({\mathbb{Z}}/2 ;{{\mathbb F}_2}) {\buildrel \delta \over \longrightarrow} H^{*+1}({\mathbb{Z}}/2; {\mathbb{Z}})$]{} (4.6,1.7)[(0,-1)[1.2]{}]{} (7.6,1.7)[(0,-1)[1.2]{}]{} (4.8,1)[$res^{{{\bf A}_4}}_{{\mathbb{Z}}/2}$]{} (7.8,1)[$res^{{{\bf A}_4}}_{{\mathbb{Z}}/2}$]{}
Let $x_1$ and $x_2$ generate $H^1({\mathbb{Z}}/2;{{\mathbb F}_2})$ and $H^2({\mathbb{Z}}/2;{\mathbb{Z}})$ respectively. From results in [@A-M], $res^{{{\bf A}_4}}_{{\mathbb{Z}}/2}$ on the left sends $u_2$ to $x_1^2$, $v_3$ to $0$, and $w_3$ to $x_1^3$. By extension, $u_2^i w_3^j$ is mapped to $x_1^{2i+3j}$ if $j$ is odd; all other classes are mapped to zero. The bottom map sends $x_1^{2i-1}$ to $x_2^i$. Since we determined the top map in Lemma \[Af\], we should be able to determine the restriction map on the right.
Consider the class $u_2^{2i+1} w_3^{2j-1}$, which maps non-trivially to $x_2^{2i+3j}$ under the composition of the left and bottom maps. $Sq^1(u_2^{2i+1} w_3^{2j-1}) = u_2^{2i} w_3^{2j-2} (u_2^3 + v_3 w_3)$, which corresponds to the class $y_4^i (y_3^2 + y_6)^{j-1} y_6$ in the integral cohomology. We claim that the only component in the expansion of this product with a non-trivial image under $res^{{{\bf A}_4}}_{{\mathbb{Z}}/2}$ is the term $y_4^i y_6^j$.
Note that if $Sq^1(z) = 0$, then $Sq^1 (w z) = (Sq^1 w)z$ for any class $w$. In $H^*({{\bf A}_4};{{\mathbb F}_2})$, $u_2^3 + v_3 w_3 = v_3^2+w_3^2$. Thus, $Sq^1 (u_2^3 + v_3 w_3) = 0$, so $Sq^1 \left( u_2^{2j+1} v_3^{i-1} (u_2^3 + v_3 w_3)^k \right) = u_2^{2j} v_3^i (u_2^3 + v_3 w_3)^k$, which corresponds to the class $y_3^i y_4^j y_6^k$. On the other hand, since any term in $H^*({{\bf A}_4};{{\mathbb F}_2})$ containing a $v_3$ maps to zero under $res^{{{\bf A}_4}}_{{\mathbb{Z}}/2}$, $y_3^i y_4^j y_6^k$ must map to $0$ also.
To finish the argument, $Sq^1 (u_2^{2i-2} w_3)$ corresponds to $y_4^i$ and maps non-trivially onto $x_2^{2i}$. Therefore, the restriction map is only non-zero when it sends $y_4^i y_6^j$ to $x_2^{2i+3j}$. [$\square$]{}
Using a similar, but easier, analysis, one can show that $res^{{{\bf D}_2}}_{{\mathbb{Z}}/2}$ sends powers of $x_2 \in H^2({{\bf D}_2}; {\mathbb{Z}})$ isomorphically to powers of the corresponding generator of $H^2({\mathbb{Z}}/2;{\mathbb{Z}})$ and vanishes otherwise. \[DtontoZ2\]
We next move to the calculation of three amalgamated products that appear as base groups for a number of the Bianchi groups presented in this article. In many cases, a Bianchi group’s cohomology is nearly identical to its base group’s, so the following results consolidate later calculations.
$H^*({{\bf D}_2}\ast_{{\mathbb{Z}}/2} {{\bf D}_2};{\mathbb{Z}})_{(2)} \cong {\mathbb{Z}}[x_2, y_2, z_2, y_3, z_3]$ with relations $y_2 z_3 = z_2 y_3 = 0$, $y_3^2 = x_2 y_2 (x_2 + y_2)$, and ${z_3}^2 = x_2 z_2 (x_2 + z_2)$. All classes have order 2. \[DtZtDt\]
[[*Proof.*]{} ]{}We use Formula (\[LES\]) with integer coefficients: $$\longrightarrow H^*({{\bf D}_2}\ast_{{\mathbb{Z}}/2} {{\bf D}_2};{\mathbb{Z}}) {\buildrel \rho \over \longrightarrow}H^*({{\bf D}_2};{\mathbb{Z}}) \oplus H^*({{\bf D}_2};{\mathbb{Z}}) {\buildrel \alpha \over \longrightarrow}H^*({\mathbb{Z}}/2;{\mathbb{Z}}) \longrightarrow$$ By Remark \[DtontoZ2\] $\alpha$ is surjective, so this sequence breaks into short exact sequences. Let $x_2,\ y_2,\ y_3$ and $x'_2,\ z_2,\ z_3$ be generators in each copy of $H^*({{\bf D}_2};{\mathbb{Z}})$ in the direct sum, and let $w_2$ be the generator of $H^2({\mathbb{Z}}/2)$ that is the restriction of both $x_2$ and $x'_2$ under $\alpha$. Now $y_2,\ z_2,\ y_3,\ z_3$ and $x_2 + x'_2$ are all in $ker (\alpha)$ and have order 2. Since this kernel is isomorphic to the image of $\rho$, we can construct the cohomology ring structure of ${{\bf D}_2}\ast_{{\mathbb{Z}}/2} {{\bf D}_2}$ at this point. We abuse notation and refer to classes in $H^*({{\bf D}_2}\ast_{{\mathbb{Z}}/2} {{\bf D}_2};{\mathbb{Z}})$ by their image under $\rho$, although we rename $x_2 + x'_2$ as $\bar x_2$. Most of the relations are obvious; note that $y_3^2 = \bar x_2 y_2 (\bar x_2 + y_2)$ as products of $x'_2$ with $y_2$ equal $0$. That the relations we found form a minimal set follows from a Poincaré series argument. We drop the bar on the $x_2$ for brevity.
One can derive the same result by using the Bockstein spectral sequence. Briefly, start with $H^*({{\bf D}_2}\ast_{{\mathbb{Z}}/2} {{\bf D}_2};{{\mathbb F}_2}) \cong {{\mathbb F}_2}[p_1, q_1, r_1] /p_1 q_1 = 0$ [@Be] and make the following identifications: $x_2 \leftrightarrow r_1^2, y_2 \leftrightarrow p_1^2, z_2 \leftrightarrow q_1^2, y_3 \leftrightarrow p_1 r_1(p_1 + r_1)$, and $z_3 \leftrightarrow q_1 r_1 (q_1 + r_1)$. These are the Bocksteins of the classes $r_1, p_1, q_1, p_1 r_1$, and $q_1 r_1$ respectively. It is straightforward to confirm that the stated relations hold. [$\square$]{}
For ${{\bf D}_2}\ast_{{\mathbb{Z}}/2} {{\bf D}_2}$, the integral calculations are possible using either the Bockstein spectral sequence or Formula \[LES\]. In practice, both techniques are often needed to determine the answer. The long exact sequence is an effective means to determine generators in low degree, and show exactly how classes arise from pieces of the amalgamated product. The Bockstein spectral sequence, on the other hand, is helpful in the determination of the relations between generators.
$H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf D}_2}; {\mathbb{Z}})_{(2)} \cong {\mathbb{Z}}[y_2, y_3, z_3, y_4, z_4, y_5, y_6]$. All classes are of order two, and satisfy the relations given in the table below. \[AfZtDt\]
[[*Proof.*]{} ]{}By results in [@Be], $H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf D}_2};{{\mathbb F}_2}) \cong {{\mathbb F}_2}[x_1,y_2,u_2,v_3,w_3]$ with relations $u_2^3 + v_3^2 + w_3^2 + v_3w_3 = 0,\ x_1^2u_2 = y_2^2,\ x_1w_3 = u_2y_2,\ y_2w_3 = x_1u_2^2$, and $x_1v_3 =
y_2v_3 = 0$. In addition, $Sq^1 (x_1) = x_1^2$, $Sq^1 (y_2) = x_1 (u_2 + y_2)$, $Sq^1 (u_2) = v_3$, and $Sq^1 (w_3) = u_2^2$.
We apply the Bockstein spectral sequence and identify generators for the integral cohomology ring. In general, one writes down classes in the mod-2 cohomology by degree and determines their Steenrod squares. Every time an integral class is encountered that is not in the ideal generated by the previous classes, it is added to the generating set. Relations among classes of order 2 can be determined using the relations in the mod-2 cohomology ring. There are some ways to speed up this occasionally difficult process. Once all integral generators have been found, it is straightforward to write down all possible combinations of generators by degree. Using the Poincaré series for the mod-2 cohomology and the Universal Coefficient Theorem, one knows how many classes are in the integral cohomology in each degree; this determines the number of relations among the classes. One continues this process until an integral Poincaré series argument or other appropriate technique shows that all generators and relations have been found.
Following this process, we identify generators $y_2 \leftrightarrow x_1^2 = Sq^1(x_1);\ y_3 \leftrightarrow x_1 (u_2 + y_2) = Sq^1(y_2);\ z_3 \leftrightarrow v_3 = Sq^1(u_2);\ y_4 \leftrightarrow x_1^2 u_2 = Sq^1(x_1 u_2);\ z_4 \leftrightarrow u_2^2 = Sq^1(w_3);\ y_5 \leftrightarrow x_1^2 w_3 + x_1 u_2^2 = Sq^1(x_1 w_3)$; and $y_6 \leftrightarrow u_2^3 + v_3 w_3 = Sq^1(u_2 w_3).$ In this example, the number of generators makes it difficult to confirm that the spectral sequence collapses at $E_2$. However, using Formula \[LES\] we get the sequence $$\rightarrow H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf D}_2};{\mathbb{Z}})_{(2)} {\buildrel \rho \over \longrightarrow}H^*({{\bf A}_4};{\mathbb{Z}})_{(2)} \oplus H^*({{\bf D}_2};{\mathbb{Z}}) {\buildrel \alpha \over \longrightarrow}H^*({\mathbb{Z}}/2;{\mathbb{Z}}) \rightarrow .$$ The map $\alpha$ is a surjection, which follows from Lemma \[AftoZ2\] and Remark \[DtontoZ2\]. Therefore, the long exact sequence breaks into short exact sequences and all 2-torsion in $H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf D}_2};{\mathbb{Z}})_{(2)}$ must be of order 2. This implies that $E_1 = E_{\infty}$. We include a table of relations in the cohomology ring, organized by degree.
Degree Relation(s)
-------- ---------------------------------------------------------------------------------------
5 $y_2z_3=0$
6 $y_3 z_3 = 0$, $y_3^2 = y_2(y_4+z_4)$
7 $z_3 y_4 = 0$, $y_3 y_4 = y_2 y_5$
8 $z_3 y_5 = 0$ , $y_4^2 = y_2^2 z_4$, $y_4 z_4 = y_2 y_6$, $y_3 y_5 = y_4^2 + y_2 y_6$
9 $y_3 y_6 = z_4 y_5$, $y_2 y_3 z_4 = y_4 y_5$
10 $y_5^2 = y_3^2 z_4$, $y_4 y_6 = y_2 z_4^2$
11 $y_5 y_6 = y_3 z_4^2$
12 $y_6^2 = y_6 z_3^2 + z_3^4 +z_4^3$
As a sample confirmation of the relations, note that $y_5^2 \leftrightarrow x_1^2 u_2^4 + x_1^4 w_3^2$ and $y_3 z_4^2 \leftrightarrow x_1^2 u_2^4 + x_1^2 u_2^2 y_2^2$. Using the relation $x_1 w_3 = u_2 y_2$ from the mod-2 ring, we find that the two expressions are equal. [$\square$]{}
$H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4}; {\mathbb{Z}})_{(2)} \cong {\mathbb{Z}}[y_3, y_4, y_6, z_6, y_7, y_9] \oplus {\mathbb{Z}}[t_3]$. All classes have order 2 with the exception of $t_3$, which has order 4. The relations are given in the table below. \[AfZtAf\]
[[*Proof.*]{} ]{}$H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4};{{\mathbb F}_2}) \cong {{\mathbb F}_2}[u_2,v_3, \bar v_3,w_3] \oplus {{\mathbb F}_2}[s_2]$ with relations $u_2^3 + w_3^2 + v_3^2 + \bar v_3^2 + w_3 (v_3 + \bar v_3) = 0,\ v_3 \bar v_3 = 0$ [@Be]. Also, $Sq^1 (u_2) = v_3 + \bar v_3$ and $Sq^1 (w_3) = u_2^2$.
We identify the following classes in the integral cohomology ring: $y_3 \leftrightarrow v_3 + \bar v_3 = Sq^1(u_2);\ y_4 \leftrightarrow u_2^2 = Sq^1(w_3);\ y_6 \leftrightarrow u_2^3 + (v_3 + \bar v_3) w_3 = Sq^1(u_2 w_3);\ z_6 \leftrightarrow v_3^2 = Sq^1(u_2 v_3);\ y_7 \leftrightarrow u_2^2 v_3 = Sq^1(v_3 w_3);$ and $y_9 \leftrightarrow v_3 (u_2^3 + v_3 w_3) = Sq^1(u_2 v_3 w_3)$. An exhaustive search finds all of the relations among these classes.
Degree Relation(s)
-------- ----------------------------------------------------------------------------------
10 $y_3 y_7 = y_4 z_6$
12 $y_3^4 + y_4^3 + y_3^2 y_6 + y_6^2 =0$, $y_3 y_9 = y_6 z_6$, $z_6^2 = y_3^2 z_6$
13 $y_6 y_7 = y_4 y_9$, $z_6 y_7 = y_3^2 y_7$
14 $y_7^2 = y_4^2 z_6$
15 $z_6 y_9 = y_3 y_6 z_6$
16 $y_7 y_9 = y_4 y_6 z_6$
18 $y_9^2 = y_3 y_6 y_9$, $y_9^2 + y_3 z_6 y_9 + y_4 y_7^2 + z_6^3 = 0$
Note that $Sq^1 (v_3) = Sq^1 \bar v_3 = 0$, whereas $Sq^1 (u_2) = v_3 + \bar v_3$. That means that there is a class in the integral cohomology which has at least order 4. We show that this class has exactly order 4 by considering the short exact sequence of 2-primary groups. $$0 {\buildrel \alpha \over \longrightarrow}H^2({\mathbb{Z}}/2;{\mathbb{Z}}) {\buildrel \delta \over \longrightarrow}H^3({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4};{\mathbb{Z}})_{(2)} {\buildrel \rho \over \longrightarrow}H^3({{\bf A}_4};{\mathbb{Z}})_{(2)} \oplus H^3({{\bf A}_4};{\mathbb{Z}})_{(2)} {\buildrel \alpha \over \longrightarrow}0.$$ Using Lemma \[Af\], we substitute to get $$0 \longrightarrow {\mathbb{Z}}/2 \longrightarrow H^3({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4};{\mathbb{Z}})_{(2)} \longrightarrow {\mathbb{Z}}/2 \oplus {\mathbb{Z}}/2 \longrightarrow 0.$$ We know from the Bockstein spectral sequence that there are two classes in $H^3({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4};{\mathbb{Z}})_{(2)}$, exactly one of which has order greater than two. This implies that $H^3({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4};{\mathbb{Z}})_{(2)} \cong {\mathbb{Z}}/2 \oplus {\mathbb{Z}}/4$. We denote the class of order 4 by $t_3$.
It remains to determine products with $t_3$. First, $t_3^2 = 0$, as $t_3$ is the only class in the cohomology ring of order 4. Without loss of generality, we can say that $\rho (t_3) = v_3$. Since $\rho (x_3) = v_3 + \bar v_3$, if $t_3 x_3$ is a non-zero product it must be $x_6$, as both classes map to $v_3^2$ under $\rho$ and there is only one such class. But $x_6^2 \neq 0$, whereas $(t_3 x_3)^2 = 0$; we conclude $t_3 x_3 = 0$. Similar arguments show that all products with $t_3$ are 0. [$\square$]{}
The Bianchi Groups ${{\Gamma}_{2}}$ and ${{\Gamma}_{6}}$
========================================================
In this section, we calculate the integral cohomology rings of ${{\Gamma}_{2}}$ and ${{\Gamma}_{6}}$ in detail. We use the Bockstein spectral sequence to determine the 2-torsion, and Formula \[LES\] for the rest of the ring structure. In here and what follows, Roman letters refer to torsion classes, and Greek letters to torsion-free classes.
$H^*(\Gamma_2;{\mathbb{Z}})_{(2)} \cong {\mathbb{Z}}[y_2, t_3, y_3, y_4, y_5, y_6, y_7, \sigma_1]$. All torsion classes have order 2, except for $t_3$ which has order 4. These classes satisfy the relations given in the table below, and $\sigma_1 y_2 = 2 t_3$; all other products with $\sigma$ and $t_3$ are trivial. $H^*({{\Gamma}_{2}};{\mathbb{Z}})_{(3)} \cong {\mathbb{Z}}[x_2, \sigma_1]$ with relations $\sigma_1^2 = 3x_2 = 0$.
[[*Proof.*]{} ]{}From [@Be], $H^*(\Gamma_2;{{\mathbb F}_2}) \cong {{\mathbb F}_2}[n_1,m_2,n_3, m_3] \oplus {{\mathbb F}_2}[s_1] \oplus {{\mathbb F}_2}[s_2]$ with relations $n_1n_3 = 0$ and $m_2^3 + m_3^2 + n_3^2 + m_3n_3 + n_1m_2m_3 = 0$. All products with ${\sigma}_1$ vanish except for the product ${\sigma}_1 n_1 = \sigma_2$. The non-trivial Bocksteins are $Sq^1 (n_1) = n_1^2$, $Sq^1 (m_2) = n_1 m_2 + n_3$, and $Sq^1 (m_3) = m_2^2$.
We apply the Bockstein spectral sequence and identify generators $y_2 \leftrightarrow n_1^2 = Sq^1(n_1);\ y_3 \leftrightarrow n_1 m_2 + n_3 = Sq^1(m_2);\ y_4 \leftrightarrow m_2^2 = Sq^1(m_3);\ y_5 \leftrightarrow n_1^2 m_3 + n_1 m_2^2 = Sq^1(n_1 m_3);\ y_6 \leftrightarrow n_1 m_2 m_3 + n_3 m_3 + m_2^3 = Sq^1(m_2 m_3)$; and $y_7 \leftrightarrow m_2^2 n_2 = Sq^1(m_3 n_3)$. The following table contains the relations by degree.
Degree Relation(s)
-------- --------------------------------------------------------------------------
8 $y_2^2 y_4 = y_2 y_3^2$, $y_2 y_6 = y_3 y_5$
9 $y_2 y_7 = 0$
10 $y_2 y_4^2 = y_3 y_7 + y_3^2 y_4$, $y_5^2 = y_2^2 y_6 + y_2 y_4^2$
11 $y_5 y_6 = y_3^2 y_5 + y_3 y_4^2 + y_4 y_7$, $y_2 y_3 y_6 = y_2 y_4 y_5$
12 $y_5 y_7 = 0$, $y_6^2 = y_3^2 y_6 + y_3^4 + y_2^2 y_4^2 + y_4^3$
13 $y_6 y_7 = y_3 y_4 y_6 + y_4^2 y_5$
14 $y_7^2 = y_2 y_4^3 + y_3^2 y_4^2$
It remains to determine what happens to the classes $s_1$ and $s_2$, and whether there are classes in the integral cohomology with order 4 or higher. We use Flöge’s presentation for the group: $\Gamma_2 =\ <A,V,S,M,U; A^2 = S^3 = (AM)^2 = M^2 = V^3 = 1,\ AM = SV^2,\ U^{-1}AU = M,\
U^{-1}SU = V>$ [@Fl]. Set $G =\ <A,V,S,M; A^2 = M^2 = (AM)^2 = S^3 = V^3 = 1,\ AM = SV^2>$ and consider the subgroups $$\begin{aligned}
G_1 & = & <A,M; A^2 = M^2 = (AM)^2 = 1>\ \cong {{\bf D}_2}, \\
G_2 & = & <S,V; S^3 = V^3 = (SV^2)^2 = 1>\ \cong {{\bf A}_4}, \\
H & = & <AM = SV^2,\ (AM)^2 = 1 >\ \cong {\mathbb{Z}}/2 . \end{aligned}$$ Let $G$ be the amalgamated product $G_1 {\ast}_H G_2 \cong {{\bf D}_2}{\ast}_{{\mathbb{Z}}/2} {{\bf A}_4}$. Then $\Gamma_2 = G \ast_{{PSL_2({\mathbb Z})}}$, where ${PSL_2({\mathbb Z})}=\ <A,S>$ and the twisting is induced by the group element $U$.
We have already calculated $H^*(G;{\mathbb{Z}})_{(2)}$ in Lemma \[AfZtDt\], but in order to keep track of the classes more carefully, we look at this example again from the point of view of group structure. The amalgamated subgroup in $G$ is $<AM>$, so let the two classes $u_2, v_2 \in H^2({{\bf D}_2};{\mathbb{Z}})$ correspond to the subgroups $<AM>$ and $<A>$. Note that $\alpha$ sends $u_2$ to the generator of $H^2({\mathbb{Z}}/2;{\mathbb{Z}})$ but sends $v_2$ to 0. The map $\alpha$ is a surjection by Lemma \[AftoZ2\] and Remark \[DtontoZ2\], so the long exact sequence splits into short exact sequences. It follows that in $H^*(G;{\mathbb{Z}})_{(2)}$ there is a single class in degree 2 and two classes in degree 3, all of order 2.
Next consider the HNN extension that generates ${{\Gamma}_{2}}$, where the group element $U$ twists $<A>$ to $<M>$. As ${PSL_2({\mathbb Z})}\cong {\mathbb{Z}}/2 \ast {\mathbb{Z}}/3$, let $w_2$ be the generator of $H^2({PSL_2({\mathbb Z})};{\mathbb{Z}})$ corresponding to the subgroup $<A>$. Then $res^{{{\Gamma}_{2}}}_{{PSL_2({\mathbb Z})}} (v_2) = \theta^*(v_2)= w_2$, so $\alpha(v_2) = 0$. We can fit this information into Formula \[LES\] for an HNN extension: $$0 \rightarrow H^2(\Gamma_2) {\buildrel \rho \over \longrightarrow}{\mathbb{Z}}/2 {\buildrel \alpha \over \longrightarrow}{\mathbb{Z}}/2 {\buildrel \delta \over \longrightarrow}H^3(\Gamma_2) {\buildrel \rho \over \longrightarrow}{\mathbb{Z}}/2 \oplus {\mathbb{Z}}/2 \rightarrow 0.$$ The map $\alpha$ is 0, so $\rho$ is an isomorphism. We abuse notation and denote the class in $H^2({{\Gamma}_{2}};{\mathbb{Z}})_{(2)}$ by $v_2$ as well. (Notice that $res^{{{\Gamma}_{2}}}_{{\mathbb{Z}}/2} (v_2) = w_2$, as both classes are dual to the group element $A$.) Since the Bockstein spectral sequence guarantees the existence of a single class of order 2 in $H^3(\Gamma_2)$, there must be another class in degree 3 of order 4, denoted by $t_3$. Also, there is a torsion-free HNN class in degree 1, $\sigma_1$. Now by Lemma \[compat\], $\sigma_1 v_2 = \delta(1) \cup v_2 = \delta(1 \cup res^{{{\Gamma}_{2}}}_{{\mathbb{Z}}/2}(v_2)) = \delta(w_2) = 2 t_3$. Finally, one can show that $\alpha$ is a surjection in degrees 3 and higher. This implies that $t_3$ is the only 2-torsion class of order greater than 2, which finishes the argument.
To determine $H^*({{\Gamma}_{2}};{\mathbb{Z}})_{(3)}$, we have that $H^*(G;{\mathbb{Z}})_{(3)} \cong H^*({\mathbb{Z}}/3;{\mathbb{Z}})$, so all that remains is to determine the effect of the final HNN extension. The group element $U$ sends $S$ to $V$, which are both elements of degree 3 in ${{\bf A}_4}$. Therefore, both the injection and twisting maps are isomorphisms, which implies that $\alpha$ is the zero map. We get short exact sequences $$0 \rightarrow H^*({PSL_2({\mathbb Z})};{\mathbb{Z}})_{(3)} {\buildrel \delta \over \longrightarrow}H^{*+1}({{\Gamma}_{2}};{\mathbb{Z}})_{(3)} {\buildrel \rho \over \longrightarrow}H^{*+1}({\mathbb{Z}}/3;{\mathbb{Z}})_{(3)} {\buildrel \alpha \over \longrightarrow}0$$ This yields classes $x_2^n \in H^{2n}({{\Gamma}_{2}};{\mathbb{Z}})_{(3)}$ and classes $\delta(w_2^n) \in H^{2n+1}({{\Gamma}_{2}};{\mathbb{Z}})_{(3)}$, where $w_2$ is the generator of $H^2({PSL_2({\mathbb Z})};{\mathbb{Z}})_{(3)}$. All classes have order 3. There is also a torsion-free HNN class, $\delta(1) \in H^1({{\Gamma}_{2}};{\mathbb{Z}})_{(3)}$, which is the class $\sigma_1$ we identified before. We claim that products with this class generate the odd dimensional cohomology. For note that $res^{{{\Gamma}_{2}}}_{{PSL_2({\mathbb Z})}} (x_2^n) = w_2^n$. Thus, $\delta(w_2^n) = \delta(w_2^n \cup 1) = \delta(res^{{{\Gamma}_{2}}}_{{PSL_2({\mathbb Z})}} (x_2^n) \cup 1) = x_2^n \cup \delta(1) = x_2^n \cup \sigma_1$. [$\square$]{}
$H^*(\Gamma_6 ; {\mathbb{Z}})_{(2)} \cong {\mathbb{Z}}[y_3, y_4, y_6, z_6, y_7, y_9] \oplus {\mathbb{Z}}[y_2, \tau_1] \oplus {\mathbb{Z}}[t_3] \oplus {\mathbb{Z}}[\sigma_1] \oplus {\mathbb{Z}}[\sigma_2]$. All torsion classes have order 2 except for $t_3$, which is of order 4. Relations among these classes are almost identical to $H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4}; {\mathbb{Z}})_{(2)}$. Products with torsion-free classes are trivial except for $y_2^k \tau_1$. At the prime 3, $H^*({{\Gamma}_{6}};{\mathbb{Z}})_{(3)} \cong {\mathbb{Z}}[x_2, \tau_1] \oplus {\mathbb{Z}}[\sigma_1] \oplus {\mathbb{Z}}[\sigma_2]$ with relations $\tau_1^2 = \sigma_1^2 = \sigma_2^2 = 3 x_2 = 0$.
[[*Proof.*]{} ]{}$H^*(\Gamma_6;{{\mathbb F}_2}) \cong {{\mathbb F}_2}[y_1, u_2,v_3, \bar v_3,w_3] \oplus {{\mathbb F}_2}[s_1] \oplus {{\mathbb F}_2}[t_1] \oplus {{\mathbb F}_2}[s_2] \oplus {{\mathbb F}_2}[t_2]$ with relations $u_2^3 + w_3^2 + v_3^2 + \bar v_3^2 + w_3 (v_3 + \bar v_3) = 0,\ v_3 \bar v_3 = 0$, and all products of other classes with each other 0 except for $y_1^j t_1$. The Bocksteins are $Sq^1(y_1) = y_1^2$, $Sq^1(u_2) = v_3 + \bar v_3$ and $Sq^1(w_3) = u_2^2$ [@Be].
As ${{\Gamma}_{6}}$ and ${{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4}$ have almost identical mod-2 cohomology, it is no surprise that most of their generators and relations in the integral cohomology are identical too; we refer the reader to Lemma \[AfZtAf\] for the definitions of $y_3, y_4, y_6, z_6, z_7$, and $y_9$, and the table of their relations. However, ${{\Gamma}_{6}}$ has a new class $y_2$ which corresponds to $y_1^2 = Sq^1(y_1)$. We switch techniques and directly calculate the cohomology of ${{\Gamma}_{6}}$ using its group structure to find relations with $y_2$.
Flöge’s presentation for $\Gamma_6$ is $<A,B,M,R,S,U,W; A^2=B^2=M^2=R^3=S^3=(BR)^3=(BS)^3=1,\ AS=MR,\ U^{-1}AU = M,\ U^{-1}SU=R,\ W^{-1}MW=A,\ W^{-1}RBW=SB>$ [@Fl]. Let $G$ be the subgroup generated by the group elements $A,B,M,R,S$ and their relations. The following subgroups of $G$, $$\begin{aligned}
G_1 & = & <S,A,B; A^2=S^3=B^2=(BS)^3=1 > \ \cong {\mathbb{Z}}/2 \ast {{\bf A}_4}, \\
G_2 & = & <R,B,M; M^2=R^3=B^2=(BR)^3=1 > \ \cong {\mathbb{Z}}/2 \ast {{\bf A}_4}, \\
H & = & <B,AS=MR; B^2 =1 > \ \cong {\mathbb{Z}}/2 \ast {{\mathbb{Z}}}, \end{aligned}$$ combine to form $G \cong G_1 {\ast}_H G_2 \cong ({\mathbb{Z}}/2 \ast {{\bf A}_4}) {\ast}_{({\mathbb{Z}}/2 \ast {\mathbb Z}) } ({\mathbb{Z}}/2 \ast {{\bf A}_4})$. The associated subgroups of the HNN extensions, $<A,\ S>$ and $<M,\ RB>$, are both isomorphic to ${PSL_2({\mathbb Z})}$. We add the group element $U$ to get the HNN extension $G_3 = G \ast_{{PSL_2({\mathbb Z})}}$. Finally, we add $W$ to get $\Gamma_6 = G_3 \ast_{{PSL_2({\mathbb Z})}}$. Thus, the cohomology of $\Gamma_6$ can be calculated in three steps.
We use Formula \[LES\] to find $H^*(G;{\mathbb{Z}})_{(2)}$. $$\ldots {\buildrel \delta \over \longrightarrow}H^*(G;{\mathbb{Z}})_{(2)} {\buildrel \rho \over \longrightarrow}H^*({\mathbb{Z}}/2 \ast {{\bf A}_4};{\mathbb{Z}})_{(2)} \oplus H^*({\mathbb{Z}}/2 \ast {{\bf A}_4};{\mathbb{Z}})_{(2)} {\buildrel \alpha \over \longrightarrow}H^*({\mathbb{Z}}/2 \ast {\mathbb{Z}};{\mathbb{Z}}) {\buildrel \delta \over \longrightarrow}\ldots.$$
From the group presentation, we see that the amalgamated copy of ${\mathbb{Z}}/2$ injects into both copies of ${{\bf A}_4}$, forming ${{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4}$ as a subgroup. The only torsion-free class in the sequence, in $H^1({\mathbb{Z}};{\mathbb{Z}})$, is not hit by $\alpha$. Its image under $\delta$ yields a torsion-free class in degree 2 which we denote by $\sigma_2$. Since this class and $2 t_3$ are the only ones in the image of $\delta$, they have trivial products with the rest of the ring elements by Lemma \[compat\] and with each other by Lemma \[PY\]. The other copies of ${\mathbb{Z}}/2$, corresponding to the group elements $A$ and $M$, go to 0 under $\alpha$. Thus, they appear as two copies of $H^*({\mathbb{Z}}/2)$ in $H^*(G;{\mathbb{Z}})_{(2)}$, which is isomorphic to the reduced direct sum $H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4};{\mathbb{Z}})_{(2)} \widetilde{\oplus} H^*({\mathbb{Z}}/2;{\mathbb{Z}}) \widetilde{\oplus} H^*({\mathbb{Z}}/2;{\mathbb{Z}}) \oplus {\mathbb{Z}}[\sigma_2]$. Let $v_2$ and $w_2$, dual to $<A>$ and $<M>$, generate the two copies of $H^2({\mathbb{Z}}/2;{\mathbb{Z}})$.
For the first HNN extension, the element $U$ twists $A$ to $M$, with $<A> \subseteq {PSL_2({\mathbb Z})}$, the associated subgroup of the extension. By Formula \[LES\], we get the sequence $$\ldots {\buildrel \delta \over \longrightarrow}H^*(G_3;{\mathbb{Z}})_{(2)} {\buildrel \rho \over \longrightarrow}H^*(G;{\mathbb{Z}})_{(2)} {\buildrel \alpha \over \longrightarrow}H^*({PSL_2({\mathbb Z})};{\mathbb{Z}})_{(2)} {\buildrel \delta \over \longrightarrow}\ldots.$$ Now as $\alpha$ is a surjection in positive degrees the long exact sequence breaks into short exact sequences. In $H^*(G;{\mathbb{Z}})_{(2)}$, the sum of the two generators $v_2 + w_2$ is in the kernel of $\alpha$ because of the twisting; let $y_2$ be the inverse image under $\rho$ of this new generator. There is also an HNN class, $\sigma_1$, which has trivial products as it is the only class that arises from the image of $\delta$. Thus, $H^*(G_3;{\mathbb{Z}})_{(2)} \cong H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4};{\mathbb{Z}})_{(2)} \oplus H^*({\mathbb{Z}}/2) \oplus {\mathbb{Z}}[\sigma_1] \oplus {\mathbb{Z}}[\sigma_2]$.
The final HNN extension yields $H^*(\Gamma_6;{\mathbb{Z}})_{(2)}.$ Here the twisting sends $M$ back to $A$. In terms of the maps induced by inclusion of twisting, $res^{{{\Gamma}_{6}}}_{{PSL_2({\mathbb Z})}} (w_2) = \theta^*(v_2)$, so $\alpha(y_2) = 0$. We abuse notation and let $y_2$ also denote its inverse image in $H^*({{\Gamma}_{6}};{\mathbb{Z}})_{(2)}$ that arises from the kernel of $\alpha$. The long exact sequence of Formula \[LES\] breaks into short exact sequences that include a degree shift: $$0 {\buildrel \alpha \over \longrightarrow}H^*({PSL_2({\mathbb Z})};{\mathbb{Z}})_{(2)} {\buildrel \delta \over \longrightarrow}H^{*+1}(\Gamma_6;{\mathbb{Z}})_{(2)} {\buildrel \rho \over \longrightarrow}H^{*+1}(G_3;{\mathbb{Z}})_{(2)} {\buildrel \alpha \over \longrightarrow}0.$$ New classes in $H^*(\Gamma;{\mathbb{Z}})_{(2)}$ include the HNN class $\tau_1 = \delta(1)$ and $\delta(u_2^j)$ in all odd degrees, where $u_2 \in H^2({PSL_2({\mathbb Z})};{\mathbb{Z}})$ corresponds to the copy of ${\mathbb{Z}}/2$ generated by the group element $M$. We use Lemma \[compat\] to show that the new classes in odd degree are actually products of other classes. Let $i$ denote the subgroup inclusion ${PSL_2({\mathbb Z})}\hookrightarrow {{\Gamma}_{6}}$. By construction, $res^{{{\Gamma}_{6}}}_{{PSL_2({\mathbb Z})}} (y_2) = u_2$. Therefore, $\delta(u_2^k) = \delta(res^{{{\Gamma}_{6}}}_{{PSL_2({\mathbb Z})}} (y_2^k) \cup 1) = y_2^k \cup \delta(1) = y_2^k \tau_1$.
The 3-primary cohomology calculation is similar. First, $H^*(G;{\mathbb{Z}})_{(3)} \cong H^*({\mathbb{Z}}/3;{\mathbb{Z}}) \widetilde{\oplus} H^*({\mathbb{Z}}/3;{\mathbb{Z}}) \oplus {\mathbb{Z}}[\sigma_2]$. The torsion-free class, $\sigma_2$, is the image of $H^1({\mathbb{Z}};{\mathbb{Z}})$ under $\delta$. The first HNN extension has the effect of identifying the two polynomial classes as it twists one copy of ${\mathbb{Z}}/3$ to the other. Thus, $H^*(G_3;{\mathbb{Z}})_{(3)} \cong H^*({\mathbb{Z}}/3;{\mathbb{Z}}) \oplus {\mathbb{Z}}[\sigma_1] \oplus {\mathbb{Z}}[\sigma_2]$, where $\sigma_1$ is the torsion-free HNN class.
For the final HNN extension, $W$ twists the second copy of ${\mathbb{Z}}/3$ back to the original one; thus $\alpha$ is 0. The long exact sequence splits as in the case of ${{\Gamma}_{2}}$, with the same results: we get a new exterior HNN class $\tau_1$, which multiplies non-trivially with a new polynomial class in degree 2. [$\square$]{}
The Bianchi Groups ${{\Gamma}_{1}}$, ${{\Gamma}_{3}}$, ${{\Gamma}_{5}}$, ${{\Gamma}_{7}}$, ${{\Gamma}_{10}}$, and ${{\Gamma}_{11}}$
====================================================================================================================================
$H^*({{\Gamma}_{1}};{\mathbb{Z}})_{(2)} \cong H^*({{\bf A}_4};{\mathbb{Z}})_{(2)} \widetilde{\oplus} H^*({{\bf D}_2};{\mathbb{Z}})_{(2)}$. $H^*({{\Gamma}_{1}};{\mathbb{Z}})_{(3)} \cong {\mathbb{Z}}[x_3, x_4]$ with relations $3 x_3 = 3 x_4 = x_3^2 = 0$.
[[*Proof.*]{} ]{}Fine shows that ${{\Gamma}_{1}}$ has the group presentation $<A, B, C, D; A^3 = B^2 = C^3 = D^2 = (AC)^2 = (AD)^2 = (BD)^2 = (BC)^2 = 1 >$ [@F]. We break this into subgroups. $$\begin{aligned}
G_{11} & = & <A,C; A^3 = C^3 = (AC)^2 = 1>\ \cong {{\bf A}_4}\\
G_{12} & = & <A,D; A^3 = D^2 = (AD)^2 = 1>\ \cong {{\bf S}_3}\\
G_{21} & = & <B,C; B^2 = C^3 = (BC)^2 = 1>\ \cong {{\bf S}_3}\\
G_{22} & = & <B,D; B^2 = D^2 = (BD)^2 = 1>\ \cong {{\bf D}_2}\end{aligned}$$ Set $G_1 = G_{11} \ast_{<A>} G_{12}$ and $G_2 = G_{21} \ast_{<B>} G_{22}$. Then $\Gamma_1$ is an amalgamated product $G_1 \ast_{{PSL_2({\mathbb Z})}} G_2$, where ${PSL_2({\mathbb Z})}\cong <C,D>$.
Using Formula \[LES\] for an amalgamated product, one easily shows that $H^*(G_1;{\mathbb{Z}})_{(2)} \cong H^*({{\bf A}_4};{\mathbb{Z}})_{(2)}$, as $\alpha$ maps $H^*({{\bf S}_3};{\mathbb{Z}})_{(2)}$ isomorphically onto $H^*({\mathbb{Z}}/2;{\mathbb{Z}})_{(2)}$. On the other hand, $H^*(G_2;{\mathbb{Z}})_{(2)} \cong H^*({{\bf D}_2};{\mathbb{Z}})_{(2)} \oplus H^*({{\bf S}_3};{\mathbb{Z}})_{(2)}$. The 2-primary result follows by noting that the copy of $H^*({{\bf S}_3};{\mathbb{Z}})_{(2)}$ maps isomorphically onto $H^*(PSL_2({\mathbb{Z}});{\mathbb{Z}})_{(2)}$ in the final amalgamated product.
For the 3-primary component, the restriction maps for ${{\bf A}_4}$ and ${{\bf S}_3}$ to ${\mathbb{Z}}/3$ (Lemmas \[St\] and \[Af\]) imply that both $H^*(G_1 ;{\mathbb{Z}})_{(3)}$ and $H^*(G_2 ;{\mathbb{Z}})_{(3)} \cong {\mathbb{Z}}[x_4]/ (3x_4 = 0)$. From Formula \[LES\] there is a long exact sequence $$\ldots H^*(\Gamma_1;{\mathbb{Z}})_{(3)} {\buildrel \rho \over \longrightarrow}H^*(G_1;{\mathbb{Z}})_{(3)} \oplus H^*(G_2;{\mathbb{Z}})_{(3)} {\buildrel \alpha \over \longrightarrow}H^*({PSL_2({\mathbb Z})};{\mathbb{Z}})_{(3)} {\buildrel \delta \over \longrightarrow}\ldots$$ Both classes in degree 4 map onto the square of the generator $w_2 \in H^2({PSL_2({\mathbb Z})};{\mathbb{Z}})_{(3)}$. Therefore, we get classes $x_4^n \in H^{4n}(\Gamma_1;{\mathbb{Z}})_{(3)}$ from the kernel of $\alpha$ and classes $\delta(w_2^{4n+2}) \in H^{4n+3}(\Gamma_1;{\mathbb{Z}})_{(3)}$. These latter classes are exterior by degree considerations. Note that $res^{{{\Gamma}_{1}}}_{{PSL_2({\mathbb Z})}}(x_4^n) = w_2^{2n}$. Thus, $\delta(w_2^{2n+1}) = \delta(res^{{{\Gamma}_{1}}}_{{PSL_2({\mathbb Z})}}(x_4^n) \cup w_2) = x_4^n \cup \delta(w_2) = x_4^n x_3$ by Lemma \[compat\]. [$\square$]{}
$H^*({{\Gamma}_{3}};{\mathbb{Z}})_{(2)} \cong H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4};{\mathbb{Z}})_{(2)}$. $H^*({{\Gamma}_{3}};{\mathbb{Z}})_{(3)} \cong {\mathbb{Z}}[x_2,x_4]$ with relations $3x_2 = 3x_4 = x_2x_4 = 0$.
[[*Proof.*]{} ]{}From calculations in [@Be], $H^*({{\Gamma}_{3}}; {{\mathbb F}_2}) \cong {{\mathbb F}_2}[u_2,\ v_3,\ \bar v_3,\ w_3] \oplus {{\mathbb F}_2}[s_2]$ with relations $u_2^3 + w_3^2 + v_3^2 + \bar v_3^2 + w_3 (v_3 + \bar v_3) = 0,\ v_3 \bar v_3 = 0$. This is identical to $H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4}; {{\mathbb F}_2})$ given in Lemma \[AfZtAf\].
The 3-primary calculations are more involved. We use the fact that ${{\Gamma}_{3}}$ acts on a retract known as the “Mendoza Complex” [@M] with fundamental domain and isotropy groups [@S-V] given in Figure 1:
(12,2.5) (5,0)[(3,4)[1.5]{}]{} (6.5,2)[(3,-4)[1.5]{}]{} (5,0)[(1,0)[3]{}]{} (4.3,0)[${{\bf A}_4}$]{} (8.3,0)[${{\bf A}_4}$]{} (6.3,2.2)[${{\bf S}_3}$]{} (6.3,0.3)[${\mathbb{Z}}/3$]{} (5.1,1.1)[${\mathbb{Z}}/2$]{} (7.4,1.1)[${\mathbb{Z}}/2$]{}
One now uses the equivariant spectral sequence given in [@Bro] to show that $H^*(\Gamma_3;{\mathbb{Z}})_{(3)} \cong H^*({{\bf A}_4};{\mathbb{Z}})_{(3)} \widetilde{\oplus} H^*({{\bf S}_3};{\mathbb{Z}})_{(3)}$. The details are identical, but easier than the those presented in [@Be]. We refer the interested reader there. [$\square$]{}
$H^*({{\Gamma}_{5}} ; {\mathbb{Z}})_{(2)} \cong {\mathbb{Z}}[y_2, z_2, y_3, y_5, z_5, y_6] \oplus {\mathbb{Z}}[t_3] \oplus {\mathbb{Z}}[\sigma_1] \oplus {\mathbb{Z}}[\tau_1] \oplus {\mathbb{Z}}[\sigma_2]$. The class $t_3$ has order 4; all other torsion classes have order two. Relations are given in the table below. $H^*({{\Gamma}_{5}};{\mathbb{Z}})_{(3)} \cong {\mathbb{Z}}[x_2, \tau_1] \oplus {\mathbb{Z}}[\sigma_1] \oplus {\mathbb{Z}}[\sigma_2]$ with relations $\tau_1^2 = \sigma_1^2 = \sigma_2^2 = 3 x_2 = 0$.
[[*Proof.*]{} ]{}From [@Be], the polynomial classes of $H^*({{\Gamma}_{5}} ; {{\mathbb F}_2})$ generate the ring ${{\mathbb F}_2}[l_1, m_1, m_3] / m_3^2 + l_1^2 m_1 m_3 + l_1 m_1^2 m_3 = 0$. There are also four exterior classes, two each in dimensions 1 and 2, which have trivial products with other classes. The Bocksteins are $Sq^1 (l_1) = l_1^2$ and $Sq^1 (m_1) = m_1^2$. Enumerating classes by degree, we find the six generators mentioned above: $y_2 \leftrightarrow l_1^2 = Sq^1 (l_1)$; $z_2 \leftrightarrow m_1^2 = Sq^1 (m_1)$; $y_3 \leftrightarrow l_1^2 m_1 + l_1 m_1^2 = Sq^1 (l_1 m_1)$; $y_5 \leftrightarrow l_1^2 m_3 = Sq^1 (l_1 m_3)$; $z_5 \leftrightarrow m_1^2 m_3 = Sq^1 (m_1^2 m_3)$; and $y_6 \leftrightarrow m_3^2 = Sq^1 (l_1 m_1 m_3)$. This last relation follows as $m_3^2 = l_1^2 m_1 m_3 + l_1 m_1^2 m_3$ in mod-2 cohomology. These six integral classes satisfy the following relations:
Degree Relation(s)
-------- ------------------------------------------
6 $y_3^2 = y_2 z_2 ( y_2 + z_2 )$
7 $y_2 z_5 = z_2 y_5$
8 $y_2 y_6 = y_3 y_5, z_2 y_6 = y_3 z_5$
10 $y_5^2 = y_2^2 y_6, z_5^2 = z_2^2 y_6$
12 $y_6^2 = y_2^2 z_2 y_6 + y_2 z_2^2 y_6 $
We use the group structure to finish the calculations. Flöge’s presentation for ${{\Gamma}_{5}}$ is $<A,B,M,R, S, U, W; A^2 = B^2 = M^2 = R^3 = S^3 = (AB)^2 = (BM)^2=1,\ AS=MR,\ U^{-1}AU = M, U^{-1}SU=R,\ W^{-1}MBW=AB,\ W^{-1}RW=S>$ [@Fl]. We write this as a double HNN extension, with base group $G$, generated by $<A,B,M,R,S>$. Consider the subgroups $$\begin{aligned}
G_1 & = & <S,A,B; S^3=A^2=B^2=(AB)^2=1 > \ \cong {\mathbb{Z}}/3 \ast {{\bf D}_2}\\
G_2 & = & <R,B,M; R^3=B^2=M^2=(MB)^2=1 > \ \cong {\mathbb{Z}}/3 \ast {{\bf D}_2}\\
H & = & <B,AS=MR; B^2 =1 > \ \cong {\mathbb{Z}}/2 \ast {{\mathbb{Z}}}. \\\end{aligned}$$ Thus $G \cong G_1 {\ast}_H G_2 \cong ({\mathbb{Z}}/3 \ast {{\bf D}_2}) {\ast}_{({\mathbb{Z}}/2 \ast {{\mathbb{Z}}}) } ({{\bf D}_2}\ast {\mathbb{Z}}/3)$. Adding the group element $U$, set $G_3 = G \ast_{{PSL_2({\mathbb Z})}}$. Adding $W$, $\Gamma_5 = G_3 \ast_{{PSL_2({\mathbb Z})}}$. Both associated subgroups, $<A,S; A^2 = S^3 = 1>$ and $<MB,R; (MB)^2 = R^3 =1>$, are isomorphic to ${PSL_2({\mathbb Z})}$.
Use Formula \[LES\] for an amalgamated product. $$\ldots {\buildrel \delta \over \longrightarrow}H^*(G;{\mathbb{Z}})_{(2)} {\buildrel \rho \over \longrightarrow}H^*({{\bf D}_2};{\mathbb{Z}})_{(2)} \oplus H^*({{\bf D}_2};{\mathbb{Z}})_{(2)} {\buildrel \alpha \over \longrightarrow}H^*({\mathbb{Z}}/2 \ast {\mathbb{Z}};{\mathbb{Z}})_{(2)} {\buildrel \delta \over \longrightarrow}\ldots.$$ By Lemma \[DtZtDt\], we see that $H^*(G;{\mathbb{Z}})_{(2)} \cong H^*({{\bf D}_2}\ast_{{\mathbb{Z}}/2} {{\bf D}_2}) \oplus {\mathbb{Z}}[\sigma_2]$; $\sigma_2$ is the image under $\delta$ of the torsion-free class in $H^1({\mathbb{Z}}/2 \ast {\mathbb{Z}};{\mathbb{Z}})_{(2)}$. Explicitly, $H^*(G;{\mathbb{Z}})_{(2)} \cong {\mathbb{Z}}[x_2, y_2, z_2, y_3, z_3] \oplus {{\mathbb F}_2}[\sigma_2]$ with appropriate relations. Assume that $y_2$ and $z_2$ are associated to $<A>$ and $<M>$ respectively.
Next we add $U$ (twisting $A$ to $M$) to get $G_3$. By Formula \[LES\], $$\ldots {\buildrel \delta \over \longrightarrow}H^*(G_3;{\mathbb{Z}})_{(2)} {\buildrel \rho \over \longrightarrow}H^*(G;{\mathbb{Z}})_{(2)} {\buildrel \alpha \over \longrightarrow}H^*({PSL_2({\mathbb Z})};{\mathbb{Z}})_{(2)} {\buildrel \delta \over \longrightarrow}\ldots.$$ Denote the HNN class by $\sigma_1$. The map $\alpha$ is surjective in positive even dimensions, and as $res^{{{\Gamma}_{3}}}_{{PSL_2({\mathbb Z})}} (y_2) = \theta^*(z_2)$, $y_2 + z_2$ is in the kernel of $\alpha$. $H^2(G_3;{\mathbb{Z}})_{(2)}$ contains two classes: $l_2=y_2+z_2$, and $m_2$, which is the preimage of $x_2$. Also, $\alpha$ is the zero map in degree 3, so $H^3(G_3;{\mathbb{Z}})_{(2)}$ contains the two classes corresponding to those in $H^3(G;{\mathbb{Z}})_{(2)}$. Finally, $H^*(G_3;{\mathbb{Z}})_{(2)}$ has a new class in degree 4 (as $\alpha (x_2 y_2) = 0$) which we denote by $z_4$.
The final calculation builds $\Gamma_5$ from $G_3$ by adding the group element $W$. This element twists a copy of ${PSL_2({\mathbb Z})}$, sending $<MB>$ to $<AB>$. Denote by $w_2$ the generator of $H^2({PSL_2({\mathbb Z})};{\mathbb{Z}})_{(2)}$. In terms of cohomology, $res^{{{\Gamma}_{5}}}_{{PSL_2({\mathbb Z})}}(x_2) = res^{{{\Gamma}_{5}}}_{{PSL_2({\mathbb Z})}}(z_2) = w_2$, and $\theta^* (x_2) = \theta^* (y_2) = w_2$. Therefore, $\alpha$ sends both $l_2$ and $m_2$ to zero, so $\delta (w_2)$ has a non-trivial image in $H^3(\Gamma_5;{\mathbb{Z}})_{(2)}$. In addition, $\alpha$ is the zero map in degree 3. Thus, there is a short exact sequence $$0 {\buildrel \alpha \over \longrightarrow}{\mathbb{Z}}/2 {\buildrel \delta \over \longrightarrow}H^3(\Gamma_5;{\mathbb{Z}})_{(2)} {\buildrel \rho \over \longrightarrow}{\mathbb{Z}}/2 \oplus {\mathbb{Z}}/2 {\buildrel \alpha \over \longrightarrow}0.$$ The calculations from the Bockstein spectral sequence imply that there is only one torsion class in degree 3 of order 2. We conclude that $H^3(\Gamma_5)$ consists of a class of order 2, and another class of order 4, which we call $t_3$. We also get an HNN class, $\tau_1$. That $t_3$ has trivial products follows from similar reasoning as in the calculation of $H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4}; {\mathbb{Z}})_{(2)}$ in Lemma \[AfZtAf\]. That $\tau_1$ has trivial products follows as it is the only class in the image of $\delta$. To finish the argument, $\alpha$ is surjective in all other degrees. For example, it is straightforward to check that $\alpha (z_4) \neq 0$. Thus, the final long exact sequence breaks into short exact sequences, which implies that all other 2-torsion classes have order 2. These have already been accounted for by the Bockstein spectral sequence, so the calculation is finished.
$H^*({{\Gamma}_{5}};{\mathbb{Z}})_{(3)}$ is calculated in an identical manner as the case ${{\Gamma}_{6}}$. [$\square$]{}
$H^*({{\Gamma}_{7}};{\mathbb{Z}})_{(2)} \cong {\mathbb{Z}}[y_2, \sigma_1]$ with relations $\sigma_1^2 = 2y_2 = 0$. $H^*({{\Gamma}_{7}};{\mathbb{Z}})_{(3)} \cong {\mathbb{Z}}[x_3,x_4] \oplus {\mathbb{Z}}[\sigma_1]$ with relations $3x_3 = 3x_4 = x_3^2 = 0$.
[[*Proof.*]{} ]{}From [@Be], $H^*({{\Gamma}_{7}};{{\mathbb F}_2}) \cong {{\mathbb F}_2}[x_1,y_1]$ with the single relation $y_1^2 = 0$. The Bockstein spectral sequence yields one torsion class in the integral cohomology, $x_2 \leftrightarrow x_1^2 = Sq^1 (x_1)$, which is polynomial. The integral class is harder to detect.
We use an HNN description due to Fine, ${{\Gamma}_{7}} \cong \left( {{\bf S}_3}\ast_{{\mathbb{Z}}/2} {{\bf S}_3}\right) \ast_{{PSL_2({\mathbb Z})}}$ [@F]. The twisting sends a ${\mathbb{Z}}/2$ in the first copy of ${{\bf S}_3}$ to the second, and a ${\mathbb{Z}}/3$ in the second copy of ${{\bf S}_3}$ back to the first. Note that $H^*({{\bf S}_3}\ast_{{\mathbb{Z}}/2} {{\bf S}_3};{\mathbb{Z}})_{(2)} \cong H^*({\mathbb{Z}}/2; {\mathbb{Z}})_{(2)}$. Furthermore, at the level of cohomology, the two dimensional generators of each copy of $H^*({{\bf S}_3};{\mathbb{Z}})_{(2)}$ are identified in the amalgamated product.
We use Formula \[LES\] for an HNN extension. $$\ldots \rightarrow H^*({{\Gamma}_{7}};{\mathbb{Z}})_{(2)} {\buildrel \rho \over \longrightarrow}H^*({{\bf S}_3}\ast_{{\mathbb{Z}}/2} {{\bf S}_3};{\mathbb{Z}})_{(2)} {\buildrel \alpha \over \longrightarrow}H^*({PSL_2({\mathbb Z})};{\mathbb{Z}})_{(2)} \rightarrow \ldots$$ The restriction and twisting maps coincide, so $\alpha = 0$. We denote the generator of $H^2({PSL_2({\mathbb Z})};{\mathbb{Z}})_{(2)}$ by $w_2$, and the generator in $H^2({{\Gamma}_{7}};{\mathbb{Z}})_{(2)}$ by $y_2$, and note that $res^{{{\Gamma}_{7}}}_{{PSL_2({\mathbb Z})}} (y_2) = w_2$. There is also the HNN class, $\sigma_1$. Substituting, the long exact sequence breaks into short exact sequences: $$0 \rightarrow H^*({\mathbb{Z}}/2;{\mathbb{Z}}) {\buildrel \delta \over \longrightarrow}H^{*+1}(\Gamma_7;{\mathbb{Z}})_{(2)} {\buildrel \rho \over \longrightarrow}H^{*+1}({{\bf S}_3}\ast_{{\mathbb{Z}}/2} {{\bf S}_3};Z)_{(2)} \rightarrow 0.$$ The classes in the odd dimensions are the products $y_2^n \sigma_1$. This follows from Lemma \[compat\], as $\delta (w_2^n) = \delta(res^{{{\Gamma}_{7}}}_{{PSL_2({\mathbb Z})}}(y_2^n) \cup 1) = y_2^n \cup \delta (1) = y_2^n \sigma_1$.
At the prime 3, $H^*({{\bf S}_3}\ast_{{\mathbb{Z}}/2} {{\bf S}_3};{\mathbb{Z}})_{(3)} \cong H^*({\mathbb{Z}}/3;{\mathbb{Z}}) \widetilde{\oplus} H^*({\mathbb{Z}}/3;{\mathbb{Z}})$. The rest of the argument is similar to the case ${{\Gamma}_{1}}$, with the exception of the HNN class. Notice that no class in $H^*({{\Gamma}_{7}};{\mathbb{Z}})_{(3)}$ restricts to the generator of $H^2({PSL_2({\mathbb Z})};{\mathbb{Z}})_{(3)}$, so all products with the HNN class are trivial. [$\square$]{}
$H^*({{\Gamma}_{10}} ; {\mathbb{Z}})_{(2)} \cong {\mathbb{Z}}[y_2, z_2, y_3, y_5, z_5, y_6] \oplus {\mathbb{Z}}[t_3] \oplus {\mathbb{Z}}[\sigma_1] \oplus {\mathbb{Z}}[\tau_1] \oplus {\mathbb{Z}}[\eta_1] \oplus {\mathbb{Z}}[\sigma_2] \oplus {\mathbb{Z}}[\tau_2]$. These satisfy the relations given in the table for ${{\Gamma}_{5}}$. All torsion classes have order 2 except for $t_3$, which has order 4. $H^*({{\Gamma}_{10}};{\mathbb{Z}}) \cong {\mathbb{Z}}[x_2, \tau_1] \oplus {\mathbb{Z}}[\sigma_1] \oplus {\mathbb{Z}}[\eta_1] \oplus {\mathbb{Z}}[\sigma_2] \oplus {\mathbb{Z}}[\tau_2]$ with relations $\sigma_2^2 = \tau_1^2 = \eta_1^2 = \sigma_2^2 = \tau_2^2 = 3 x_2 = 0$.
[[*Proof.*]{} ]{}The statement about the torsion follows as the polynomial classes in the mod-2 cohomology are identical to the case ${{\Gamma}_{5}}$. The only difference is that $H^*({{\Gamma}_{10}};{{\mathbb F}_2})$ contains three exterior classes in each of degrees 1 and 2 that have trivial products with all other classes [@Be].
Flöge’s presentation for ${{\Gamma}_{10}}$ is $<A,B,L,S,D,U,W;\ A^2=B^2=L^2=S^3=(AB)^2=(AL)^2=1>$ with other relations involving $D,\ U$, and $W$ that we give in their respective extensions [@Fl]. The base group $G_0$ has the presentation $<A,B,L,S;\ A^2=B^2=L^2=S^3=(AB)^2=(AL)^2=1>$ which we break into pieces. $$\begin{aligned}
G_{01} & = & <A,B; A^2 = (AB)^2 = B^2 = 1 > \ \cong {{\bf D}_2}, \\
G_{02} & = & <A,L; A^2 = (AL)^2 = L^2 = 1 > \ \cong {{\bf D}_2}, \\
G_{03} & = & <S> \ \cong {\mathbb{Z}}/3, \\
H & = & <A> \ \cong {\mathbb{Z}}/2. \end{aligned}$$
With this decomposition, $G_0 \cong (G_{01} {\ast}_H\ G_{02})\ {\ast}\ G_{03} \cong ({{\bf D}_2}{\ast}_{{\mathbb{Z}}/2} {{\bf D}_2}) \ast {\mathbb{Z}}/3$. The first HNN extension, $G_1$, adds the element $D$. Its presentation is $<G_0, D; D^{-1}ALSD = S^{-1}AB>$, with $<ALS> \ \cong {\mathbb Z}$. $G_2$, the second HNN extension, adds the group element $U$. Explicitly, $G_2 = \ <G_1, U;
U^{-1}DABD^{-1}U = D^{-1}ALD, U^{-1}LDS^{-1}D^{-1}U = BD^{-1}S^{-1}D>$. The associated subgroup $<DABD^{-1}, LDS^{-1}D^{-1}>$ is isomorphic to ${\mathbb{Z}}/2 \ast {{\mathbb{Z}}}$. The final HNN extension is ${{\Gamma}_{10}} = \ <G_2,W; W^{-1}BW = U^{-1}LU,
W^{-1}D^{-1}SDW = U^{-1}DSD^{-1}U>$, where the associated subgroup $<B, D^{-1}SD>$ is isomorphic to ${PSL_2({\mathbb Z})}$.
At the prime 2, $H^*(G_0;{\mathbb{Z}})_{(2)} \cong H^*({{\bf D}_2}\ast_{{\mathbb{Z}}/2} {{\bf D}_2};{\mathbb{Z}})_{(2)}$. For the first HNN extension we use Formula \[LES\]. $$\ldots {\buildrel \delta \over \longrightarrow}H^*(G_1;{\mathbb{Z}})_{(2)} {\buildrel \rho \over \longrightarrow}H^*(G_0;{\mathbb{Z}})_{(2)} {\buildrel \alpha \over \longrightarrow}H^*({\mathbb{Z}}) {\buildrel \delta \over \longrightarrow}\ldots .$$ This implies that $H^*(G_1;{\mathbb{Z}})_{(2)} \cong H^*(G_0;{\mathbb{Z}})_{(2)}$ with two exceptions: there are new torsion-free classes in degrees 1 and 2. All products with these classes vanish by Lemmas \[compat\] and \[PY\].
The next two HNN extensions proceed in a similar manner as the case ${{\Gamma}_{5}}$. These calculations account for the class $t_3$, and each HNN extension generates an HNN class in degree 1. The second HNN extension also generates a torsion-free class in degree two, the image of $\delta H^1({\mathbb{Z}}\ast {\mathbb{Z}}/2;{\mathbb{Z}})$. Therefore, $H^*({{\Gamma}_{10}};{\mathbb{Z}})_{(2)}$ has the same torsion classes as in the case of ${{\Gamma}_{5}}$, as well as three torsion-free classes in degree 1 and two in degree 2.
In fact, $H^*({{\Gamma}_{10}};{\mathbb{Z}})_{(3)}$ is also close to $H^*({{\Gamma}_{5}};{\mathbb{Z}})_{(2)}$. Note that $H^*(G_0;{\mathbb{Z}})_{(3)} \cong H^*({\mathbb{Z}}/3;{\mathbb{Z}})$. These are the only torsion classes in $H^*(G_1;{\mathbb{Z}})_{(3)}$ and $H^*(G_0;{\mathbb{Z}})_{(3)}$, as these HNN extensions twist copies of ${\mathbb{Z}}$ and ${\mathbb{Z}}\ast {\mathbb{Z}}/2$. The HNN extensions also give rise to two torsion-free classes in each of the degrees 1 and 2, the classes $\sigma_1$, $\sigma_2$, $\tau_1$, and $\tau_2$.
It remains to determine the effect of the final HNN extension, where a copy of ${PSL_2({\mathbb Z})}$ is twisted. The original copy of ${\mathbb{Z}}/3$ was generated by the group element $S$. Although it is not twisted by $W$, a conjugate of it is, which induces an isomorphism on cohomology. Therefore, we see that both the restriction and the twisting maps are isomorphisms onto $H^*({\mathbb{Z}}/3;{\mathbb{Z}})$, which implies $\alpha = 0$. The long exact sequence breaks into short exact sequences, as in the case of ${{\Gamma}_{5}}$. We get an HNN class, $\eta_1 = \delta(1)$, and this class multiplies non-trivially with powers of the generator in $H^2({{\Gamma}_{10}};{\mathbb{Z}})_{(3)}$ that arises from $\alpha$ being the zero map. [$\square$]{}
$H^*({{\Gamma}_{11}};{\mathbb{Z}})_{(2)} \cong H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4};{\mathbb{Z}})_{(2)} \oplus {\mathbb{Z}}[\sigma_1].$ The class $\sigma_1$ is torsion-free. $H^*({{\Gamma}_{11}};{\mathbb{Z}})_{(3)} \cong {\mathbb{Z}}[x_2, \sigma_1]$ with relations $\sigma_1^2 = 3x_2 = 0$.
[[*Proof.*]{} ]{}The Bockstein spectral sequence generates almost the entire integral ring structure from the mod-2 cohomology of the group, which is $H^*(\Gamma_{11};{{\mathbb F}_2}) \cong {{\mathbb F}_2}[u_2,\ v_3,\ \bar v_3,\ w_3] \oplus {{\mathbb F}_2}[s_1] \oplus {{\mathbb F}_2}[s_2]$ with relations $u_2^3 + w_3^2 + v_3^2 + \bar v_3^2 + w_3 (v_3 + \bar v_3) = 0,\ v_3 \bar v_3 = 0$. With the exception of the class $\sigma_1$, this is identical to $H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/2} {{\bf A}_4};{{\mathbb F}_2})$. The only class not accounted for is $\sigma_1$.
The alternate method of calculation resolves this issue. Fine shows in [@F] that $\Gamma_{11} \cong \left( {{\bf A}_4}{\ast}_{{\mathbb{Z}}/3} {{\bf A}_4}\right) \ast_{{PSL_2({\mathbb Z})}}$, where the twisting sends copies of ${\mathbb{Z}}/2$ and ${\mathbb{Z}}/3$ from one ${{\bf A}_4}$ to the other. Thus, there is a long exact sequence, $$\ldots \rightarrow H^*(\Gamma_{11};{\mathbb{Z}})_{(2)} {\buildrel \rho \over \longrightarrow}H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/3} {{\bf A}_4};{\mathbb{Z}})_{(2)} {\buildrel \alpha \over \longrightarrow}H^*({PSL_2({\mathbb Z})};{\mathbb{Z}})_{(2)} \rightarrow \ldots ,$$ with $H^*({PSL_2({\mathbb Z})};{\mathbb{Z}})_{(2)} \cong H^*({\mathbb{Z}}/2;{\mathbb{Z}})_{(2)}$. From this sequence we see that $\sigma_1$ is a torsion-free HNN class and the only class to arise from the image of $\delta$, so all products with $\sigma_1$ vanish. $H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/3} {{\bf A}_4};{\mathbb{Z}})_{(2)} \cong H^*({{\bf A}_4};{\mathbb{Z}})_{(2)} \oplus H^*({{\bf A}_4};{\mathbb{Z}})_{(2)}$. The result follows.
For the 3-primary cohomology, start with $H^*({{\bf A}_4}\ast_{{\mathbb{Z}}/3} {{\bf A}_4};{\mathbb{Z}})_{(3)} \cong H^*({\mathbb{Z}}/3;{\mathbb{Z}})$. The twisting sends ${\mathbb{Z}}/3$ in one copy of ${{\bf A}_4}$ to the other copy of ${{\bf A}_4}$; when one calculates the cohomology of the HNN extension, this implies that $\alpha$ is zero. We now proceed as in the case ${{\Gamma}_{2}}$. [$\square$]{}
Since the ring of integers ${{\mathcal O}}_d$ contains number theoretic information, one might hope that some of this information would be detected in the cohomology ring of the corresponding Bianchi group ${{\Gamma}_{d}}$. We note, however, that the torsion classes in $H^*({{\Gamma}_{5}};{\mathbb{Z}})$ and $H^*({{\Gamma}_{10}};{\mathbb{Z}})$ are identical–the difference is in the torsion-free classes! It is not clear at this point if there is a connection, but if there is, it is clear that many more examples will need to be calculated to find it.
[99]{}
A. Adem and J. Milgram, *Cohomology of Finite Groups*, Springer Verlag, 1994.
E. Berkove, [*On the Mod-2 Cohomology of the Bianchi Groups*]{}, Trans. of the AMS., [**352**]{}, no. 10, (2000), 4585-4602.
W. Browder [*Torsion in H-spaces*]{}, Ann. of Math, (2) [**74**]{}, 1961, 24-51.
K. Brown, *Cohomology of Groups*, Springer Verlag, 1993.
B. Fine, *Algebraic Theory of the [B]{}ianchi Groups*, Marcel Dekker, Inc. (Monographs and textbooks in pure and applied mathematics; 129), 1989.
D. Flöge, *Zur Struktur der $PSL_2$ über einigen imaginär-quadratischen Zahlringen*, Math. Zeitschrift **183** (1983), 255–279.
F. Grunewald and J. Schwermer, *Subgroups of Bianchi Groups and Arithmetic Quotients of Hyperbolic 3-Space*, Trans. of the AMS (1) **335** (Jan. 1993), 47–78.
J. McCleary, *A User’s Guide to Spectral Sequences*, Second Edition, Cambridge University Press (Cambridge Studies in advance mathematics; 58), 2001.
E. Mendoza, *Cohomology of $PGL_2$ Over Imaginary Quadratic Integers*, Bonner Math. Schriften, vol. 128, 1980.
K. Pearson, *Integral Cohomology and Detection of $\omega$-Basic 2-Groups*, Math. of Computation, Vol. 65 **213**, (Jan. 1996), 291-306.
J. Pakianathan and E. Yalcin *On G-Equivariant Coverings*, Preprint.
J. Schwermer and K. Vogtmann, *The Integral Homology of $SL_2$ and $PSL_2$ of Euclidean Imaginary Quadratic Integers*, Comment. Math. Helvetica **58** (1983), 573–598.
J. P. Serre, *Trees*, Springer-Verlag, 1980.
E. Spanier, *Algebraic Topology*, 1st reprinted version, Springer-Verlag, 1989.
K. Vogtmann, *Rational Cohomology of Bianchi Groups*, Math. Ann. **272** (1985), 399–419.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
WTC7 seems to be a classic controlled demolition. WTC 1 &2 destruction appears to have been enhanced by thermate (a variation of thermite) in addition.
Pentagon was not struck by a passenger aircraft. It was a drone or missle.
"Georgian Interior Minister Vano Merabishvili accused the Russian military Monday of being behind a series of attacks, including a powerful car bomb attack that killed three policemen in the eastern Georgian city of Gori in February. "
Osama bin Laden tried to buy a massive amount of cocaine, spike it with poison and sell it in the United States, hoping to kill thousands of Americans one year after the 9/11 attacks, the NEW YORK POST reported on Tuesday.
Please send along anything you think is important if youfeel so inclined.
What a strange time to be alive.I think it's a real privaledge.We're lucky to see glimses intothe truth's & lies at this point in the history ofhumanity.I only hope that good will win out. Hard to know.
I was in tube bomb carriage - and survived"CAMBRIDGE dancer Bruce Lait has spoken of his miraculous escape when a bomb exploded just yards away from him in a Tube train carriage.The 32-year-old was knocked out by the blast and awoke to a terrible scene of devastation in the underground tunnel near London's Aldgate East station.So lucky: Bruce recovers in hospitalMr Lait, who teaches dance in Cambridge, believes he and his dance partner Crystal Main were the only passengers in the carriage who survived the blast without serious injury - even though they were sitting nearest to where the bomb detonated.When he came to, there was a body lying on top of him and he was surrounded by the dead and injured. But incredibly, the only wounds the dance coach sustained were facial lacerations and a perforated eardrum."I feel extremely, extremely lucky," he said.The explosion happened just after Mr Lait and Ms Main, 23, got on the train at Liverpool Street on their way to the South Bank for a rehearsal.He recalled that the carriage had about 20-25 people in it, from all walks of life, and aged from their teens to over 60."I remember an Asian guy, there was a white guy with tracksuit trousers and a baseball cap, and there were two old ladies sitting opposite me," he said."We'd been on there for a minute at most and then something happened. It was like a huge electricity surge which knocked us out and burst our eardrums. I can still hear that sound now," he said.The impact of the blast made him pass out. As he came to, he wondered whether he was alive or dead."We were right in the carriage where the bomb was. I was knocked out. I did not know what was going on."I wondered if I was dead or not. I said to myself, you can't be dead because your brain is having conscious thoughts, so concentrate hard. I was telling myself 'wake up Bruce, wake up'."Royal visit: At Royal London HospitalDisorientated, he only gradually realised where he was and what had happened."When I woke up and looked around I saw darkness, smoke and wreckage. It took a while to realise where I was and what was going on, then my first concern was for Crystal."She was okay but she was in shock because she was trying to deal with the person on top of her who had massive head injuries. We have just found out that this person died," said Mr Lait, who lives in Suffolk.He too was afraid to move because there was a seriously injured woman lying on top of him."I realised someone was lying on top of me. I tried not to move her because I didn't know if she was still alive, or I could have made it worse. This person also died, while on top of me."At the same time, he slowly tried to work out whether he or Crystal had been injured."I thought if I can wiggle my toes I'm okay, and I could, and I asked Crystal to do the same."Describing the scene as they waited for help, he said: "It was just the most awful scene of death and there were body parts everywhere. There was something next to me. I was trying not to look. I couldn't figure out what it was."When paramedics arrived, they confirmed that the woman on top of him was dead and carefully moved her body. Mr Lait said the middle-aged woman had blonde curly hair, was dressed in black, and could have been a businesswoman.He and Crystal were helped out of the carriage. As they made their way out, a policeman pointed out where the bomb had been.It was like a huge electricity surge which knocked us out and burst our eardrums.Tube survivor Bruce Lait"The policeman said 'mind that hole, that's where the bomb was'. The metal was pushed upwards as if the bomb was underneath the train. They seem to think the bomb was left in a bag, but I don't remember anybody being where the bomb was, or any bag," he said.They were led through the tunnel to the platform at Aldgate, which was just a few hundred yards away, and taken out of the station to wait for an ambulance.Mr Lait was taken to the Royal London Hospital, Whitechapel, where he was visited by the Queen on Friday.He said: "They asked would I mind if my name were put forward and I said I'd love to meet the Queen, even if the circumstances weren't ideal."Sitting with his parents, Pat and Tom, Mr Lait told the Queen as she stood at his bedside: "I'm very thankful to still be here."He said of Her Majesty: "She just seemed very nice and concerned, she seemed very genuine."Now back at home, he has been trying to recover from the ordeal, with the help of friends and family.Mr Lait, who teaches the Latin formation team XS, based in Cambridge, and the Cambridge Dancers' Club, said he has been moved by people's care and consideration.On the floor: Bruce & dance partner Crystal, who was also on the train"I've had people who know me phone me from all over the world and ask if I am alright. Those pictures of me and the Queen have gone all over the world."And he said the terrible experience has given him a new outlook on life."It has made me realise how important life is, and that we only get one life, and we've got to be happy with what we've got in our lives."Reflecting on the ordeal, he said: "Out of that whole carriage, I think Crystal and I were the only ones who were not seriously injured, and I think we were nearest the bomb."It makes me thank Him up there. I'm not overly religious but I'm not a disbeliever. I pray now and again. Something like this has just made me think, 'thank you Lord'."
"We believe they are willing to sometimes sacrifice their own people for the objective to control the whole world," said Abu Hamza, president of the Islamic Information and Services Network.
Mr Hamza, who leads 250 Muslims at a Coburg mosque, said Sheik Mohammed Omran was not alone in believing the US Government was behind the 2001 attacks on New York and Washington that killed almost 3000 people.
Sheik Omran, a fundamentalist leader at the Ahlus Sunnah wal Jam'ah Association in Brunswick, has been criticised by Prime Minister John Howard for not strongly denouncing terrorism.
Sheik Omran, while saying terrorism was evil, has said the Bush Administration orchestrated the September 11 attacks to give the US licence to invade Islamic countries.
Sheik Omran also believes there is no proof al-Qaida leader Osama bin Laden has anything to do with terrorism.
Mr Hamza agreed with the conspiracy theory, saying most Australian Muslims also believed the US was to blame.
"There are many Muslims, and I am one of those Muslims, that believe whatever Omran has said is correct.
"The majority of Muslims believe; they believe that even if Muslims have done it they have been given access to do this."
He said the US now had "a licence to go into Islamic countries . . . next it will be Syria, Saudi (Arabia)".
Mr Hamza said he and other Muslims condemned terrorist attacks, even if they were carried out by Muslims.
"(But) I can't believe that Muslims can do this."
He said the Koran forbade the killing of innocent people.
"John Howard said last week the Muslim leaders are not doing enough . . . what we fear is that this could create an atrocity between Muslims and non-Muslims."
Mr Hamza said creating a view that all Muslims supported terror was akin to saying that because one Christian priest was a pedophile then all priests were pedophiles.
Islamic Friendship Association spokesman Keysar Trad denied most Muslims saw a US conspiracy behind the 2001 attacks, but said some were deeply suspicious of involvement.
"Many don't care about the conspiracy theory," Mr Trad said.
"The rest of them say whoever was behind it, if it was bin Laden, he must have been able to buy the support of some traitors within America.
"It could be anyone, it could be the CIA. There is this theory he (bin Laden) has always been working for the Americans and this was just another mission.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Twilight's Last Gleaming (novel)
Twilight's Last Gleaming is a 2014 novel by John Michael Greer. The novel is a work of speculative political fiction, in which a declining United States attempts to gain control of some desperately needed oil by overthrowing the government of Tanzania, prompting an intervention by Tanzania's ally, the People's Republic of China, with disastrous consequences.
Plot
In the year 2024, significant oil reserves are discovered in the Indian Ocean, in Tanzanian territorial waters. The following year, U.S. President Jameson Weed and his staff meet in the White House. The United States of the year 2025 was suffering from a long period of economic stagnation, exacerbated by the effects of peak oil and global warming. Weed's plan to take control of the Tanzanian oil gets support from Vice President Leonard Gurney, a political hack, and from National Security Adviser Ellen Harbin, a hawkish neocon, but is opposed by Secretary of Defense Bill Stedman, who believes the plan is asking for trouble. Overconfident and unable to conceive of an American military defeat, the Weed administration decides to launch the operation. Meanwhile, the President of Tanzania, knowing that the United States has a history of fighting for oil, senses danger and appeals to his country's ally, China, for protection.
The men of the Chinese government's inner circle, knowing that they have reached military parity with the declining United States, devise an ambitious plan to help their client state while delivering a blow to American military prestige, and they receive eager assistance from America's numerous enemies. The Chinese are able to keep their plan secret, and thus the American forces are taken by surprise when they launch the invasion in July 2025. The Chinese disable the American satellite network through a cyberattack, severely hampering American command and control, then launch hundreds of high-speed computer-guided cruise missiles which they had smuggled into Tanzania. The American fleet is not prepared to defend itself against this new generation of missiles. The Americans shoot down many of them, but as one missile after another finds its way through the ships' air defenses, the fleet is devastated: most of the ships, including the aircraft carrier USS Ronald Reagan, are either sunk or disabled.
Meanwhile, the Army and Air Force launch a land invasion from Kenya, the only American client state remaining in the region. The unexpected arrival of PLAAF aircraft (transported covertly via Russia and Iran) forces the Americans on the defensive. The American F-35 fighters were selected for their ability to produce profits for their politically connected manufacturers rather than their efficacy in combat; as a result, the Chinese military is able to gain air supremacy. Tanzania and several other African nations declare war on Kenya. The international press chronicles the fighting (reporters from Australia and the United Arab Emirates play a prominent role) but the Weed administration tries to keep the news out of American newspapers. As the Chinese strategists had anticipated, the Americans are unable to imagine that they might actually be defeated, and therefore double down on their existing military strategy, further depleting American reserves and matériel.
The Chinese and their African allies drive the American and Kenyan forces back into Kenya. With so many ships and aircraft already lost, the Americans cannot easily resupply them. The Weed administration escalates the conflict by attacking a Chinese military base directly. Aghast, Stedman resigns as Secretary of Defense. The Chinese retaliate by launching a surprise attack on the American base at Diego Garcia, taking control of the island. By September, the government of Kenya had fallen, and the American forces are trapped and must surrender. Iran, with promises of a quid pro quo from the Chinese for stirring up trouble, launches an invasion of Saudi Arabia, and the Americans are powerless to intervene.
Weed escalates the situation further, threatening a limited tactical nuclear strike on the Chinese. The President of Russia responds by promising an all-out nuclear war if the United States acts on its threat. A nuclear crisis unfolds. When local authorities fire on panicked civilians trying to flee the city of Trenton, New Jersey, several other states, led by Texas, openly defy the federal government, implementing their own plans for surviving the crisis, including the possibility of secession. Seeing his own government crumbling and no sign of concessions from Russia or China, Weed finally backs down and sues for peace.
Although the terms of the peace treaty only required the United States to accept the loss of Kenya and Diego Garcia, the defeat and the ongoing war between Iran and Saudi Arabia cause the price of imported oil, already high, to skyrocket, and the United States faces an energy crisis. Knowing that his navy is obsolete and U.S. troops in the Persian Gulf are vulnerable, Weed evacuates them, effectively shutting the U.S. out of Africa, the Middle East, and the Indian Ocean. As investors around the world realize that the United States was no longer a hyperpower, they begin selling U.S. dollar-denominated investments, causing the dollar to fall sharply. This leads to high inflation and a major economic crisis in the United States. The problem becomes worse when several large investment banks, who had taken unhedged risks that had been intended to pay off when the United States won the war, became insolvent. With the dollar in free fall, the federal government cannot bail out the banks without risking a total collapse of the dollar, and so they fall into bankruptcy, and the United States enters an economic depression. Weed resigns the presidency and commits suicide.
The new president, Gurney, inherits Weed's cabinet, including Harbin, with whom he has begun an extramarital affair. He appoints Senator Pete Bridgeport as his new Vice President, as an incentive to protect the administration from congressional hearings on the military misadventure that had ruined the country. The hearings proceed anyway. Gurney's recovery plan involves a grandiose rearmament project that the United States can no longer afford, and a domestic spending plan based almost entirely on unfunded mandates. The plan is universally condemned by the states and the people, and the state government of Arkansas passes a resolution calling for a constitutional convention to ban unfunded mandates.
Harbin, trying to protect the administration, arranges to have Stedman, the former Secretary of Defense, assassinated to prevent him from testifying before Congress. However, Stedman had been aware of this possibility, leaving instructions for his computer-whiz grandson to leak his documents in case anything happens to him. The government falls into a crisis of legitimacy, with anti-government rhetoric and donations to extremist political parties reaching new records. Other states join the call for a constitutional convention, and the convention opens in September 2026 in St. Louis, Missouri.
At the convention, a resolution calling for a constitutional amendment to ban unfunded mandates passes easily, but the delegates cannot agree on anything else. After several weeks of deadlock, a proposal to dissolve the United States of America outright surfaces. The shocked delegates pass the resolution, and the convention closes.
In Washington, Gurney and Harbin draw up plans to avert dissolution by means of a military coup, drawing up a long list of people to be purged, including Bridgeport and many other high-ranking government and military officials. The crisis comes to a head in October, when Stedman's grandson, still hacking government computers, discovers the plot and leaks it, alerting Bridgeport. As president of the Senate, Vice President Bridgeport seeks military protection from the Chairman of the Joint Chiefs of Staff, and convenes an emergency meeting of Congress. Gurney is impeached, convicted, and removed from office in one day. Gurney flees the country and is killed by unknown assailants in Brazil, while Harbin is imprisoned.
Bridgeport becomes President, and although he enjoys personal popularity and a reputation for honesty, he cannot overcome the near-universal perception that the American government is broken beyond repair. As the dollar has lost more than 90% of its value over the past year, Bridgeport recalls all American troops home from overseas, and tries to work with Congress to balance the budget with tax revenues a tiny fraction of what they had been before the war. It is too little, too late, however; numerous states begin to ratify the dissolution amendment, starting with South Carolina. In January 2027, Nebraska becomes the thirty-eighth state to ratify, and the United States of America officially ceases to exist.
In the following weeks, the states begin to arrange themselves into a number of successor states. In Washington, Bridgeport and other former government officials contemplate founding a rump United States out of states that voted against ratification, based in the Rust Belt and Mid-Atlantic regions.
Speculations
Some of the speculations that Greer made in this novel:
The United States remained dependent on imported oil. All attempts to reduce this dependence, including domestic fracking, biodiesel, solar power and others, had either failed or remained economically unfeasible. Peak oil is a common theme of Greer's work.
The United States had participated in wars in Ukraine and Venezuela in the years before the novel is set.
Tanzania had become an oil exporter before the discovery of the offshore oil field. In 2015, Tanzania was not a significant oil producer.
India is still a nuclear state, but Pakistan is not, implying that Pakistan disarmed at some point.
India and Brazil have been granted permanent seats on a reformed United Nations Security Council.
Prince William, Duke of Cambridge had become King of United Kingdom. In 2015, William was second in line to the throne after his father, Prince Charles.
The name of the military operation, Blazing Torch, was supposedly named after the national flag of Tanzania. This design was actually based on the flag of Zaire (now the Democratic Republic of the Congo), not Tanzania. This implies that the president's staffers were so overconfident or ignorant that they could not even name their plans right, although this is not addressed in the novel.
The president of Russia bore a personal grudge against the United States, blaming it for his father's suicide during the Russian economic depression of the 1990s.
Automation is mentioned briefly, but mass technological unemployment has not occurred.
The distinction between America's major political parties is portrayed as irrelevant. None of the major politicians' political affiliations are ever specified.
Production
Greer published a short version of this story on his blog, the Archdruid Report, in October 2012.
References
Category:2014 American novels
Category:Peak oil books
Category:Fiction set in 2024
Category:Fiction set in 2025
Category:Fiction set in 2026
Category:Fiction set in 2027
Category:Political fiction
Category:United States in fiction
Category:Kenya in fiction
Category:China in fiction
Category:Tanzania in fiction
Category:Novels set in Kenya
Category:Novels set in Tanzania
Category:Novels set in the Indian Ocean
Category:Novels set in Washington, D.C.
Category:Novels set in Missouri
Category:American war novels
Category:American political novels
Category:American speculative fiction novels
Category:Fictional wars
Category:Invasions in fiction
| null |
minipile
|
NaturalLanguage
|
mit
| null |
* The state shut down Tinley Park Mental Health Center on June 30, 2012. But the state kept paying a lease on its postage meter for almost another year, according to a new Auditor General’s report. That shows the state probably didn’t have its act together when it closed the facility.
During a review of 220 equipment items reported to the Department of Central Management Services as lost, valued at $157,656, the auditors noted the following:
- For equipment items with either an acquisition cost in excess of $500 or deemed a high theft, Tinley Park reported 86 items were lost, valued at $123,772. This represented 10% of the total inventory dollar value and 13% of the total number of items on hand at Tinley Park’s closure date of July 1, 2012. […]
- Tinley Park reported it had lost a refrigerated van trailer, valued at $29,217, during its equipment inventory following the closure. [Emphasis added.]
The auditors identified equipment items, including furniture, 4 farm tractors, a forklift, and 10 inoperable State vehicles that were still on site within various buildings across the Tinley Park campus and had, presumably, been abandoned. The inoperable State vehicles appeared they had been inoperable for a considerable length of time. In addition, the auditors identified one vehicle that had been abandoned by a private person at Tinley Park after it was sold by the DCMS several years ago.
The auditors also identified one operable vehicle at Tinley Park that was in use by the Department of Central Management Services’ caretaker at Tinley Park. This vehicle was still on the Tinley Park equipment inventory at June 30, 2013, totaling $18,135.
At the Jacksonville Developmental Center, which closed on November 27, 2012, auditors found confidential information in trash receptacles, filing cabinets, binders, boxes, and on desks. We found resident’s names, health information, and social security numbers, a labeled medical specimen, photos of residents’ labeled with residents’ names and incident number, security reports which included residents’ names; filing cabinets with folder separators labeled with residents’ names; two computer monitors and three computer towers; large stack of binders which contained the last name and first initial of residents on the spine of the binders; and manila filing folders with patient names written on the tabs.
I don’t believe most of the losses were theft. Sounds like a lot of the assets were old worn-out junk, too far gone to be worth the cost of removing them. And only the CMS accountants were referring to them at their original value, now long since depreciated and amortized. CMS math is…. special. And when you close a facility and lay off the staff, who is left responsible for unloading that stuff? Nobody. It’s still state property and state land until it gets sold off. So it is “abandoned in place.” Becomes the next owner’s problem.
But the personal records, that’s a bigger issue. That should have been the staff’s final job assignment on their last week: transferring and disposing of records. You can blame management for not following up on that.
Yeah Anon, not state incompetence all, just a simple misunderstanding…
Lets enjoy some other misunderstandings from the report…
Page 76
428 of 510 (84%) pharmaceutical items were inaccurately counted during the June 30, 2013 physical count. The auditors performed test counts and identified a significant number of uncounted items…
Page 83
The Department of Human Services did not adequately restrict access to its production data….
We found several instances where programmers were allowed inappropriate access to production data sets.
Page 84
From July to October Department staff sent numerous unprotected emails to our staff that contained infomration such as:
This is one of the things that drove me away from the Democratic Party.
Back in the 1970s and 1980s when I pretty much voted Dem when I lived in the 19th Ward, the general understanding was that political patronage was a way of life. What kept the local pols in power, however, was delivering services properly and satisfying the constituents. Patronage, at least in the 19th,was limted to the politically connected who at least had what it took to get the job done.
Heaven help the Precinct Captain who didn’t get the potholes filled and garbage picked up.
Then things changed. There was little pride in getting the job done by the patronage employees,and more and more grossly incompetent and recalcitrant folks started being hired over those who could do the job, and it showed.
School building engineers who kept their boiler rooms so clean that you could eat off them were replaced by folks who were more concerned with breaking their well manicured nails than getting the job done.
Like a virus it spread through the City, Cook County,and then, when Dem Governors took over, to the state.
When local Dems ran things well, I supported them. As these cases show, that’s not the case anymore. It’s time to break up this system and go back to the “old” dem system of patronage with performance, or maybe some Republicans making that happen.
The way the GOP is Illinois is run, not much chance there. If we had a few folks like Rep Franks empowered in the Dem castle, maybe things couls change, but he’s certainly an outlier in the party at this point.
I just don’t see much interest on the part of the electorate to make things work in the city, county and state anymore. Most people like me feel hopelessness with this system and if they have other options, they take them.
Do any of you insiders out there see ANY anger over this increasing waste and incompetence out there in the precincts? I sure didn’t.
I have visited several shuttered state operated facility buildings when doing consulting work for what was then the Department of Mental Health in the 90’s. Even then the state or agency lacked the skill or willingness to appropriately shutter a building and dispose of equipment and records. I encountered some pretty awful scenarios.
I have been through the closure and relocation of a state facility. It is a nightmare and the folks being moved out/ laid off are not very helpful. There are not enough people at CMS, or any agency, to move material. The procurement for hiring or contracting for people to come in and move items to a CMS warehouse is difficult at best. Most of the machinery, vehicles, equipment, costs more to move and store than they are worth. These problems will continue as the state workforce gets smaller and smaller.
The Political appointee will never lose a days pay or get fired! They are politically protected. The Union workers are protected by the Union and never lose any pay. I remember a former Deputy Director of another department saying many times! “Its not coming out of my pocket, so who cares.” Until the State and federal Government, this includes elected officials are held responsible. This will never stop and corruption, waste and theft will be rampant.
They should have a liquidation sale when these places close. They’d collect more money, they’d be rid of the excess junk, and a responsible vendor would actually provide an inventory of everything kept, sold or scrapped. Have you seen the crew that does this work? Quite a picture.
==Lets hear one more time how all the waste and corruption are gone.==
Strawman.
==I have been through the closure and relocation of a state facility. It is a nightmare and the folks being moved out/ laid off are not very helpful==
Why should those being kicked to the curb be helpful? If the state were “run like a business,” these workers wouldn’t be expected to clean up anything but their desk and they’d be out fast and there would be a cleanup crew afterwards tossing things in the trash.
There should be accountability for this up the chain in DHS but we all know that won’t happen. It’s not like they had no experience closing centers before this. This was no oversight. It was sloppy just like the closures.
That they found personal photos of residents in the trash at Jacksonville gives credence to the stories of residents being hastily moved out of the facility as it neared its closure date.
These are not the signs of thoughtful, organized transfers. These are the signs of people grabbing what they can before the sheriff shows up to lock the doors.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Enantioselective 2-hydroxylation of RS-8359, a selective and reversible MAO-A inhibitor, by cytochrome P450 in mouse and rat liver microsomes.
RS-8359, (+/-)-4-(4-cyanoanilino)-5,6-dihydro-7-hydroxy-7H-cyclopenta[d]-pyrimidine is a racemic compound with a selective and reversible monoamine oxidase A (MAO-A) inhibition activity. The substrate and product enantioselectivity with respect to 2-hydroxylation of RS-8359 enantiomers was studied using mouse and rat liver microsomes. In mice, the (S)-enantiomer was transformed to the cis-diol metabolite, whereas the (R)-enantiomer to the trans-diol metabolite. The Vmax/Km value for the formation of the cis-diol metabolite from the (S)-enantiomer was sevenfold greater than that for the formation of the trans-diol metabolite from the (R)-enantiomer. The greater Vmax/Km value for the (S)-enantiomer was due to the tenfold smaller Km value compared to that for the (R)-enantiomer. The results were in fair agreement with the previously reported low plasma concentrations of the (S)-enantiomer and the high recovery of the cis-diol metabolite derived from the (S)-enantiomer in urine after oral administration of RS-8359 to mice. Similarly to mice, in rats the (R)-enantiomer was transformed to the trans-diol metabolite, whereas the (S)-enantiomer yielded the cis-diol and trans-diol metabolites. The Vmax/Km value for the (R)-enantiomer was larger than that for the (S)-enantiomer in rats, indicating that the low plasma concentration of the (S)-enantiomer in rats might be caused by a metabolic reaction other than P450-dependent hydroxylation. CYP3A was shown to be responsible for the trans-diol formation from the (R)-enantiomer.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Q:
Using Boto3 to create loop on specific folder
I am testing the new data feeds as in XML. Those data will be stored in S3 in the following format:
2018\1\2\1.xml
2018\1\3\1.xml
2018\1\3\2.xml
etc. So, multiple .xml files are possible on one day. Also, important to note that there are folders in this bucket that I do NOT want to pull. So I have to target a very specific directory.
There is no date time stamp within the file, so I need to use created, modified, something to go off of. To do this I think of using a dictionary of key, values with folder+xml file as the key, created/modified timestamp as the value. Then, use that dict to essentially re-pull all the objects.
Here's what I've tried..
i
mport boto3
from pprint import pprint
client = boto3.client('s3')
paginator = client.get_paginator('list_objects_v2')
result = paginator.paginate(
Bucket='bucket',
Prefix='folder/folder1/folder2')
bucket_object_list = []
for page in result:
pprint(page)
if "Contents" in page:
for key in page[ "Contents" ]:
keyString = key[ "Key" ]
pprint(keyString)
bucket_object_list.append(keyString)
s3 = boto3.resource('s3')
obj = s3.Object('bucket','bucket_object_list')
obj.get()["Contents"].read().decode('utf-8')
pprint(obj.get())
sys.exit()
This is throwing an error from the key within the obj = s3.Object('cluster','key') line.
Traceback (most recent call last):
File "s3test2.py", line 25, in <module>
obj = s3.Object('cluster', key)
NameError: name 'key' is not defined
The Maxitems is purely for testing purposes although it's interesting since this translates to 1000 when run.
A:
NameError: name 'key' is not defined
As far as error is concerned, it's because key is not defined.
From this documentation:
Object(bucket_name, key)
Creates a Object resource.:
object = s3.Object('bucket_name','key')
Parameters
bucket_name(string) -- The Object's bucket_name identifier. This must be set.
key(string) -- The Object's key identifier. This must be set.
You need to assign an object key name to the 'key' you're using in the code
The keyName is the "name" (=unique identifier) by which your file will be stored in the S3 bucket
Code based on what you posted:
import boto3
client = boto3.client('s3')
paginator = client.get_paginator('list_objects_v2')
result = paginator.paginate( Bucket='bucket_name', Prefix='folder/folder1/folder2')
bucket_object_list = []
for page in result:
if "Contents" in page:
for key in page[ "Contents" ]:
keyString = key[ "Key" ]
print(keyString)
bucket_object_list.append(keyString)
print bucket_object_list
s3 = boto3.resource('s3')
for file_name in bucket_object_list:
obj = s3.Object('bucket_name',file_name)
print(obj.get())
print(obj.get()["Body"].read().decode('utf-8'))
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Short-term electrogeometric atrial remodelling after percutaneous atrial septal defect closure.
Atrial arrhythmias are well known long-term complications of atrial septal defect (ASD), presumably due to chronic atrial enlargement and stretch. Surgical repair often fails to revert the arrhythmic risk despite atrial volumetric unloading, maybe as a consequence of atriotomic scar. Avoiding atrial incision, percutaneous ASD closure should result in atrial unloading and arrhythmic risk decrease. This study evaluated the atrial short-term electrogeometric effects of percutaneous ASD closure. Fifteen asymptomatic patients (age 23 +/- 5 years) submitted to percutaneous closure of large ASD (QP/QS 2.4 +/- 0.3) using the Amplatzer Septal Occluder device (24 +/- 2 mm) underwent atrial echocardiographic (indexed right atrial size and left atrial size as well as right atrial/left atrial volume ratio) and electrocardiographic (P-wave duration and dispersion, PR conduction time and dispersion) evaluation at 1 and 6 months after ASD closure. After 6 months, the indexed right atrial volume and the right atrial/left atrial volume ratio had significantly decreased (from 39 +/- 5 to 20 +/- 2 ml/m2, P < 0.001 and from 2.0 +/- 0.2 to 1.0 +/- 0.1, P < 0.0001, respectively) as did the P-wave dispersion (from 32 +/- 2 to 28 +/- 1 ms, P = 0.03) despite a significant increase of both P-wave maximal (from 75 +/- 3 to 81 +/- 3 ms, P < 0.01) and minimal (from 42 +/- 3 to 53 +/- 3 ms, P < 0.005) duration. Percutaneous ASD closure causes a short-term positive atrial electrogeometric remodelling. The electrocardiographic predictors of atrial arrhythmias, however, tend to worsen early after device implantation despite a marked volumetric unloading, possibly due to a 'foreign body' effect of the occluding device. This might warrant a closer follow-up during the first few weeks after device implantation.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
In vivo microdialysis for the investigation of drug levels in the dermis and the effect of barrier perturbation on cutaneous drug penetration. Studies in hairless rats and human subjects.
The thesis opens with review chapters concerning theoretical and practical aspects of the investigation of drug contents in the skin. A discussion of the advantages and limitations of the established methods as well as the relatively new sampling method of microdialysis, which is employed in the experimental section, is given. Factors influencing the barrier function of the normal human skin are described as are the alterations in skin barrier function found in diseased and experimentally barrier perturbed skin. The microdialysis technique consists of introducing an ultra thin, semipermeable tube, a so-called probe, in the dermis. The tube is connected to a precision pump, which provides a steady flow of a tissue-compatible fluid through the probe at a very low flow. Smaller molecules in the tissue, among them the non-protein bound fraction of the drug content in the extracellular fluid, will passively diffuse across the surface of the membrane and thus enter the flow of the perfusate, which is sampled at regular intervals and analysed. Microdialysis is used for the determination of drug levels in the skin after topical as well as systemic drug delivery in the experimental part of the thesis. The method is not applicable to the investigation of all drugs or compounds, as we have shown that it is not feasible to sample highly protein-bound drugs or very lipophilic drugs by microdialysis without further development of the method. The investigation of topical drug administration consists of 2 studies of cutaneous penetration of a model drug, salicylic acid, initially investigated in hairless rats and subsequently in human volunteers. In both studies, barrier perturbation of the skin was undertaken by physical (removal of the stratum corneum by repeated tape stripping) or chemical (treatment with acetone) methods or by provocation of irritative dermatitis (by application of sodium lauryl sulphate, a detergent). Prior to the penetration experiment, the barrier damage inflicted was quantified by non-invasive measurements of transepidermal, water loss and erythema. The penetration of salicylic acid, applied in an ethanol solution in chambers glued to the skin in the barrier perturbed areas, was measured by microdialysis sampling of the drug level in the underlying dermis. At the end of the experiment, probe depth in the dermis and skin thickness were measured by ultrasound scanning. In humans and hairless rats alike, the cutaneous drug penetration was highly increased in tape stripped skin (157- and 170-fold increased, respectively, in comparison to the penetration in unmodified skin) and in skin with irritative dermatitis (46- and 80-fold increased). Delipidization by acetone led to a doubling of the penetration in humans but had no effect on penetration in hairless rats. In both studies a close correlation between the measurements of barrier perturbation by non-invasive methods and the cutaneous drug penetration in the same area was found. In the human study, the barrier perturbation in the acetone treated area was not measurable by non-invasive methods, whereas drug penetration, measured by microdialysis sampling, was significantly increased, indicating that the microdialysis method possesses high sensitivity in the detection and quantification of perturbed skin barrier function. In the human study, a dose-response relationship between the concentration of detergent used for the induction of irritant dermatitis and the ensuing increase in drug penetration across the skin could be demonstrated. In the hairless rat study a correlation between probe depth in the dermis and drug penetration was found, demonstrating that the more superficially a probe was placed, the earlier it would be reached by the influx of drug across the skin. Systemic drug distribution was studied in healthy volunteers following oral administration of 2 g acetylsalicylic acid. (ABSTRACT TRUNCATED)
| null |
minipile
|
NaturalLanguage
|
mit
| null |
AN/MPQ-33/39 Continuous Wave Illumination Radar (CWIR)
This
X-BandCW System
is used to illuminate targets in the Hawk Missile Battery (Basic Hawk).
The unit comes mounted on its own mobile trailer. Unit automatically acquires and tracks designated targets in azimuth elevation and range rate.
The system has an output power of around 125 W operating in the 10-10.25 GHz band.
The outer structure (except the antenna group) corresponds to
CWAR- acquisition radar.
MPQ-39 was an upgraded version of the MPQ-33.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Exemplary embodiments relate generally to wireless communications, and more particularly, to methods, systems, and computer program products for implementing a locator service.
Wireless technologies have grown in popularity for a variety of applications. For example, in the personal consumer market, wireless home networking devices provide configurable internetworking solutions for various types of home devices such as communications, computing, and entertainment devices.
On a larger scale, wireless technologies such as global satellite communications offer global positioning services for mobile devices. For example, GPS services provide mapping and direction assistance to travelers. Global positioning services are also utilized to track the location of vehicles in an effort to minimize theft. Another popular market relating to global satellite technology is the satellite radio and programming industry. Many vehicles are now equipped with wireless receivers that pick up satellite music and programming from all over the world (e.g., services provided by XM Satellite Radio, Inc. of Washington, D.C. as well as SIRIUS Satellite Radio of New York City, N.Y.). These types of applications typically involve a subscription service to a service provider.
In addition to personal consumer applications, business applications relating to wireless technologies have also enjoyed great advancements (e.g., wireless area networks, cellular communications for field activities, etc.).
As wireless technologies continue to advance, consumers, business entities, government, military, and other organizations will continue to look for ways to exploit them.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Q:
Numerical reference encoding for Perl
I'm having an issue with Perl. I'm trying to use HTML::Entities to encode a string which it does. However I'm trying to convert it into a numerical reference. e.g:
use strict;
use warnings;
use LWP::Protocol::https;
use HTML::Entities;
my $str = "vis-à-vis Beyoncé's naïve\npapier-mâché résumé";
$str = encode_entities($str);
print $str;
this out puts
vis-à-vis Beyoncé's naï
papier-mâché résumé
Instead I would like to output
vis-à-vis beyoncè naïve\npapier-mâchï rïsumï";
The issue here is I don't wanna use this type of encoding (à) but rather use this type (à).
A:
I have a simple solution using regex, to get what you need, here's the code below:
use HTML::Entities qw[encode_entities_numeric];
use utf8;
my $str = "vis-à-vis Beyoncé's naïve\npapier-mâché résumé";
$str = encode_entities_numeric($str);
print "\n=======\n",$str,"\n=======\n";
$str =~ s/&#x([A-F0-9]+);/sprintf('&#%s;',hex($1))/ge;
print "\n=======\n",$str,"\n=======\n";
The output looks something like this:
=======
vis-à-vis Beyoncé's naïve
papier-mâché résumé
=======
=======
vis-à-vis Beyoncé's naïve
papier-mâché résumé
=======
Find the pattern of the hex code, and convert it to the decimal format.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
---
abstract: 'Neural networks with attention have proven effective for many natural language processing tasks. In this paper, we develop attention mechanisms for uncertainty detection. In particular, we generalize standardly used attention mechanisms by introducing *external attention* and *sequence-preserving attention*. These novel architectures differ from standard approaches in that they use external resources to compute attention weights and preserve sequence information. We compare them to other configurations along different dimensions of attention. Our novel architectures set the new state of the art on a Wikipedia benchmark dataset and perform similar to the state-of-the-art model on a biomedical benchmark which uses a large set of linguistic features.'
author:
- Heike Adel
- |
Hinrich Schütze\
Center for Information and Language Processing (CIS)\
LMU Munich, Germany\
[[email protected]]{}\
bibliography:
- 'eacl2017.bib'
title: Exploring Different Dimensions of Attention for Uncertainty Detection
---
Introduction
============
For many natural language processing (NLP) tasks, it is essential to distinguish uncertain (non-factual) from certain (factual) information. Such tasks include information extraction, question answering, medical information retrieval, opinion detection, sentiment analysis [@karttunen; @vincze; @cruz2015] and knowledge base population (KBP). In KBP, we need to distinguish, e.g., “X may be Basque” and “X was rumored to be Basque” (uncertain) from “X is Basque” (certain) to decide whether to add the fact “Basque(X)” to a knowledge base. In this paper, we use the term *uncertain information* to refer to speculation, opinion, vagueness and ambiguity. We focus our experiments on the uncertainty detection (UD) dataset from the CoNLL2010 hedge cue detection task [@CoNLLsharedTask]. It consists of two medium-sized corpora from different domains (Wikipedia and biomedical) that allow us to run a large number of comparative experiments with different neural networks and exhaustively investigate different dimensions of attention.
Convolutional and recurrent neural networks (CNNs and RNNs) perform well on many NLP tasks [@cw; @kalchbr; @zeng2014; @rnnRelClass]. CNNs are most often used with pooling. More recently, attention mechanisms have been successfully integrated into CNNs and RNNs [@attention0; @attention1; @attention2; @attention3; @attention5; @attention6; @abcnn]. Both pooling and attention can be thought of as *selection mechanisms* that help the network focus on the most relevant parts of a layer, either an input or a hidden layer. This is especially beneficial for long input sequences, e.g., long sentences or entire documents. We apply CNNs and RNNs to uncertainty detection and compare them to a number of baselines. We show that attention-based CNNs and RNNs are effective for uncertainty detection. On a Wikipedia benchmark, we improve the state of the art by more than 3.5 $F_1$ points.
Despite the success of attention in prior work, the design space of related network architectures has not been fully explored. In this paper, we develop novel ways to calculate attention weights and integrate them into neural networks. Our models are motivated by the characteristics of the uncertainty task, yet they are also a first attempt to systematize the design space of attention. In this paper, we begin with investigating three dimensions of this space: weighted vs. unweighted selection, sequence-agnostic vs. sequence-preserving selection, and internal vs. external attention.
---------------------------------------------------------------------- ---------------------------------------------------------------------- -------------------------------------------------------------------- -----------------------------------------------------------------------------
(1) (2) (3) (4)
{height="\myfigheight\textheight"} {height="\myfigheight\textheight"} {height="\myfigheight\textheight"} {height="\myfigheight\textheight"}
---------------------------------------------------------------------- ---------------------------------------------------------------------- -------------------------------------------------------------------- -----------------------------------------------------------------------------
**Weighted vs. Unweighted Selection.** Pooling is unweighted selection: it outputs the selected values as is. In contrast, attention can be thought of as weighted selection: some input elements are highly weighted, others receive weights close to zero and are thereby effectively not selected. The advantage of weighted selection is that the model learns to decide based on the input how many values it should select. Pooling either selects all values (average pooling) or k values (k-max pooling). If there are more than k uncertainty cues in a sentence, pooling is not able to focus on all of them.
**Sequence-agnostic vs. Sequence-preserving Selection.** K-max pooling [@kalchbr] is sequence-preserving: it takes a long sequence as input and outputs a subsequence whose members are in the same order as in the original sequence. In contrast, attention is generally implemented as a weighted average of the input vectors. That means that all ordering information is lost and cannot be recovered by the next layer. As an alternative, we present and evaluate new sequence-preserving ways of attention. For uncertainty detection, this might help distinguishing phrases like “it is not uncertain that X is Basque” and “it is uncertain that X is not Basque”.
**Internal vs. External Attention.** Prior work calculates attention weights based on the input or hidden layers of the neural network. We call this internal attention. For uncertainty detection, it can be beneficial to give the model a lexicon of seed cue words or phrases. Thus, we provide the network with additional information to bear on identifying and summarizing features. This can simplify the training process by guiding the model to recognizing uncertainty cues. We call this external attention and show that it improves performance for uncertainty detection.
Previous work on attention and pooling has only considered a small number of the possible configurations along those dimensions of attention. However, the internal/external and un/weighted distinctions can potentially impact performance because external resources add information that can be critical for good performance and because weighting increases the flexibility and expressivity of neural network models. Also, word order is often critical for meaning and is therefore an important feature in NLP. Although our models are motivated by the characteristics of uncertainty detection, they could be useful for other NLP tasks as well.
Our main contributions are as follows. (i) We extend the design space of selection mechanisms for neural networks and conduct an extensive set of experiments testing various configurations along several dimensions of that space, including novel sequence-preserving and external attention mechanisms. (ii) To our knowledge, we are the first to apply convolutional and recurrent neural networks to uncertainty detection. We demonstrate the effectiveness of the proposed attention architectures for this task and set the new state of the art on a Wikipedia benchmark dataset. (iii) We publicly release our code for future research.[^1]
Models
======
**Convolutional Neural Networks.** CNNs have been successful for many NLP tasks since convolution and pooling can detect key features independent of their position in the sentence. Moreover, they can take advantage of word embeddings and their characteristics. Both properties are also essential for uncertainty detection since we need to detect cue phrases that can occur anywhere in the sentence; and since some notion of similarity improves performance if a cue phrase in the test data did not occur in the training data, but is similar to one that did. The CNN we use in this paper has one convolutional layer, 3-max pooling (see ), a fully connected hidden layer and a logistic output unit.
**Recurrent Neural Networks.** Different types of RNNs have been applied widely to NLP tasks, including language modeling [@rnnlm01; @rnnlm02], machine translation [@gatedRNN; @attention0], relation classification [@rnnRelClass] and entailment [@attention3]. In this paper, we apply a bi-directional gated RNN (GRU) with gradient clipping and a logistic output unit. showed that GRUs and LSTMs have similar performance, but GRUs are more efficient in training. The hidden layer $h$ of the GRU is parameterized by two matrices $W$ and $U$ and four additional matrices $W_r$, $U_r$ and $W_z$, $U_z$ for the reset gate $r$ and the update gate $z$ [@gatedRNN]: $$\begin{aligned}
r = \sigma(W_rx + U_rh^{t-1})\\
z = \sigma(W_zx + U_zh^{t-1})\\
h^t = z \odot h^{t-1} + (1-z)\odot \tilde{h}^{t}\\
\tilde{h}^t = \sigma(Wx + U(r \odot h^{t-1})) \eqlabel{reset}\end{aligned}$$ $t$ is the index for the current time step, $\odot$ is element-wise multiplication and $\sigma$ is the sigmoid.
Attention
=========
Architecture of the Attention Layer
-----------------------------------
We first define an attention layer $a$ for input $x$: $$\begin{aligned}
\alpha_i &= &\frac{\exp(f(x_i))}{\sum_j \exp(f(x_j))} \eqlabel{gating2}\eqlabel{attweight}\\
a_i &= & \alpha_i \cdot x_i\eqlabel{gating1}\eqlabel{attreweighted}\end{aligned}$$ where $f$ is a scoring function, the $\alpha_i$ are the attention weights and each input $x_i$ is reweighted by its corresponding attention weight $\alpha_i$.
The most basic definition of $f$ is as a linear scoring function on the input $x$: $$f(x_i) = W^Tx_i \eqlabel{internalatt}$$ $W$ are parameters that are learned in training.
{height=".08\textheight"} {height=".08\textheight"}
Focus and Source of Attention
-----------------------------
In this paper, we distinguish between focus and source of attention.
The *focus* of attention is the layer of the network that is reweighted by attention weights, corresponding to $x$ in . We consider two options for the application in uncertainty detection as shown in : (i) the focus is on the input, i.e., the matrix of word vectors ((1) and (3)) and (ii) the focus is on the convolutional layer of the CNN or the hidden layers of the RNN ((2) and (4)). For focus on the input, we apply tanh to the word vectors (see part (1) of figure) to improve results.
The *source* of attention is the information source that is used to compute the attention weights, corresponding to the input of $f$ in .
formalizes the case in which focus and source are identical (both are based only on $x$). We call this **internal attention** (see left part of ). *An attention layer is called internal if both focus and source are based only on information internally available to the network (through input or hidden layers).*[^2]
If we conceptualize attention in terms of source and focus, then a question that arises is whether we can make it more powerful by *increasing the scope of the source beyond the input*.
In this paper, we propose a way of expanding the source of attention by making an *external resource* $C$ available to the scoring function $f$: $$f(x_i) = f'(x_i,C)
\label{externalattgen}$$ We call this **external attention** (see right part of ). *An attention layer is called external if its source includes an external resource.*
The specific external-attention scoring function we use for uncertainty detection is parametrized by $U_1$, $U_2$ and $V$ and defined as follows: $$f(x_i) = \sum_j V^T \cdot \tanh(U_1 \cdot x_i + U_2 \cdot
c_j) \label{externalatt}$$ where $c_j$ is a vector representing a cue phrase $j$ of the training set. We compute $c_j$ as the average of the embeddings of the constituent words of $j$.
This attention layer scores an input word $x_i$ by comparing it with each cue vector $c_j$ and summing the results. The comparison is done using a fully connected hidden layer. Its weights $U_1$, $U_2$ and $V$ are learned during training. When using this scoring function in , each $\alpha_i$ is an assessment of how important $x_i$ is for uncertainty detection, taking into account our knowledge about cue phrases. Since we use embeddings to represent words and cues, uncertainty-indicating phrases that did not occur in training, but are similar to training cue phrases can also be recognized.
We use this novel attention mechanism for uncertainty detection, but it is also applicable to other tasks and domains as long as there is a set of vectors available that is analogous to our $c_j$ vectors, i.e., vectors that model relevance of embeddings to the task at hand (for an outlook, see Section \[outlook\]).
Sequence-agnostic vs. Sequence-preserving Selection
---------------------------------------------------
So far, we have explained the basic architecture of an attention layer: computing attention weights and reweighting the input. We now turn to the *integration of the attention layer* into the overall network architecture, i.e., how it is connected to downstream components.
The most frequently used downstream connection of the attention layer is to take the **average**: $$a = \sum_i a_i
\eqlabel{tradAttention}$$ We call this the average, not the sum, because the $\alpha_i$ are normalized to sum to 1 and the standard term for this is “weighted average”.
A variant is the **k-max average**: $$a = \sum_{R(\alpha_j)\leq k} a_j$$ where $R(\alpha_j)$ is the rank of $\alpha_j$ in the list of activation weights $\alpha_i$ in descending order. This type of averaging is more similar to k-max pooling and may be more robust because elements with low weights (which may just be noise) will be ignored.
Averaging destroys order information that may be needed for NLP sequence classification tasks. Therefore, we also investigate a sequence-preserving method, **k-max sequence**: $$a = [a_j | R(\alpha_j)\leq k]$$ where $[a_j | P(a_j)]$ denotes the subsequence of sequence $A=[a_1,
\ldots ,a_J]$ from which members not satisfying predicate $P$ have been removed. Note that sequence $a$ is in the original order of the input, i.e., not sorted by value.
K-max sequence selects a subsequence of input vectors. Our last integration method is **k-max pooling**. It ranks each dimension of the vectors individually, thus the resulting values can stem from different input positions. This is the same as standard k-max pooling in CNNs except that each vector element in $a_j$ has been weighted (by its attention weight $\alpha_j$), whereas in standard k-max pooling it is considered as is. Below, we also refer to k-max sequence as “per-pos” and to k-max pooling as “per-dim” to clearly distinguish it from k-max pooling done by the CNN.
#### Combination with CNN and RNN Output.
Another question is whether we combine the attention result with the result of the convolutional or recurrent layer of the network. Since k-max pooling (CNN) and recurrent hidden layers with gates (RNN) have strengths complementary to attention, we experiment with concatenating the attention information to the neural sentence representations. The final hidden layer then has this form: $$h = \mbox{tanh}(W_1 a + W_2 r + b)$$ with $r$ being either the CNN pooling result or the last hidden state of the RNN (see ).
{width=".3\textwidth"}
Experimental Setup and Results {#experiments}
==============================
Task and Setup
--------------
We evaluate on the two corpora of the CoNLL2010 hedge cue detection task [@CoNLLsharedTask]: Wikipedia (11,111 sentences in train, 9634 in test) and Biomedical (14,541 train, 5003 test). It is a binary sentence classification task. For each sentence, the model has to decide whether it contains uncertain information.
For hyperparameter tuning, we split the training set into core-train (80%) and dev (20%) sets; see appendix for hyperparameter values. We use 400 dimensional word2vec [@mikolov] embeddings, pretrained on Wikipedia, with a special embedding for unknown words.
For evaluation, we apply the official shared task measure: $F_1$ of the uncertain class.
Baselines without Attention
---------------------------
Our baselines are a support vector machine (SVM) and two standard neural networks without attention, an RNN and a CNN. The SVM is a reimplementation of the top ranked system on Wikipedia in the CoNLL-2010 shared task [@georgescul], with parameters set to ’s values; it uses bag-of-word (BOW) vectors that only include hedge cues. Our reimplementation is slightly better than the published result: 62.01 vs. 60.20 on wiki, 78.64 vs. 78.50 on bio.
Model wiki bio
---- -------------- -------------- --------------
() Baseline SVM 62.01$\star$ 78.64$\star$
() Baseline RNN 59.82$\star$ **84.69**
() Baseline CNN **64.94** 84.23
: $F_1$ results for UD. Baseline models without attention. $\star$ indicates significantly worse than best model (in bold).
Model wiki bio
----- -------------------- -------------- -----------
() Baseline RNN 59.82$\star$ 84.69
(4) RNN attention-only 62.02$\star$ **85.32**
(5) RNN combined 58.96$\star$ 84.88
() Baseline CNN 64.94$\star$ 84.23
(6) CNN attention-only 53.44$\star$ 82.85
(7) CNN combined **66.49** 84.69
: $F_1$ results for UD. Attention-only vs. combined architectures. Sequence-agnostic weighted average for attention. $\star$ indicates significantly worse than best model (bold).
The results of the baselines are given in . The CNN (line ) outperforms the SVM (line ) on both datasets, presumably because it considers all words in the sentence – instead of only predefined hedge cues – and makes effective use of this additional information. The RNN (line ) performs better than the SVM and CNN on biomedical data, but worse on Wikipedia. In Section \[CNNvsRNN\], we investigate possible reasons for that.
Experiments with Attention Mechanisms {#sec:attentionExperiments}
-------------------------------------
For the first experiments of this subsection, we use the sequence-agnostic weighted average for attention (see ), the standard in prior work.
**Attention-only vs. Combined Architecture.** For the case of internal attention, we first remove the final pre-output layer of the standard RNN and the standard CNN to evaluate attention-only architectures. This architecture works well for RNNs but not for CNNs. The CNNs achieve better results when the pooling output (unweighted selection) is combined with the attention output (weighted selection). See for $F_1$ scores.
The baseline RNN has the difficult task of remembering the entire sentence over long distances – the attention mechanism makes this task much easier. In contrast, the baseline CNN already has an effective mechanism for focusing on the key parts of the sentence: k-max pooling. Replacing k-max pooling with attention decreases the performance in this setup.
Since our main goal is to explore the benefits of adding attention to existing architectures (as opposed to developing attention-only architectures), we keep the standard pre-output layer of RNNs and CNNs in the remaining experiments and combine it with the attention layer as in .
**Focus and Source of Attention.** We distinguish different focuses and sources of attention. For focus, we investigate two possibilities: the input to the network, i.e., word embeddings (F=W); or the hidden representations of the RNN or CNN (F=H). For source, we compare internal (S=I) and external attention (S=E). This gives rise to four configurations: (i) internal attention with focus on the first layer of the standard RNN/CNN (S=I, F=H), see lines (5) and (7) in , (ii) internal attention with focus on the input (S=I, F=W), (iii) external attention on the first layer of RNN/CNN (S=E, F=H) and (iv) external attention on the input (S=E, F=W). The results are provided in .
For both RNN (8) and CNN (13), the best result is obtained by focusing attention directly on the word embeddings.[^3] These results suggest that it is best to optimize the attention mechanism directly on the input, so that information can be extracted that is complementary to the information extracted by a standard RNN/CNN.
For focus on input (F=W), external attention (13) is significantly better than internal attention (11) for CNNs. Thus, by designing an architectural element – external attention – that makes it easier to identify hedge cue properties of words, the learning problem is apparently made easier.
For the RNN and F=W, external attention (10) is not better than internal attention (8): results are roughly tied for bio and wiki. Perhaps the combination of the external resource and the more indirect representation of the entire sentence produced by the RNN is difficult. In contrast, hedge cue patterns identified by convolutional filters of the CNN can be evaluated well based on external attention; e.g., if there is strong external-attention evidence for uncertainty, then the effect of a hedge cue pattern (hypothesized by a convolutional filter) on the final decision can be boosted.
In summary, the CNN with external attention achieves the best results overall. It is significantly better than the standard CNN that uses only pooling, both on Wikipedia and biomedical texts. This demonstrates that the CNN can make effective use of external information – a lexicon of uncertainty cues in our case.
Model S F wiki bio
------ -------------- --- --- -------------- --------------
() Baseline RNN - - 59.82$\star$ 84.69
(5) RNN combined I H 58.96$\star$ 84.88
(8) RNN combined I W 62.18$\star$ 84.81
(9) RNN combined E H 61.19$\star$ 84.62
(10) RNN combined E W 61.87$\star$ 84.41
() Baseline CNN - - 64.94$\star$ 84.23$\star$
(7) CNN combined I H 66.49 84.69
(11) CNN combined I W 65.13$\star$ 84.99
(12) CNN combined E H 64.14$\star$ 84.73
(13) CNN combined E W **67.08** **85.57**
: $F_1$ results for UD. Focus (F) and source (S) of attention: Internal (I) vs external (E) attention; attention on word embeddings (W) vs. on hidden layers (H). Sequence-agnostic weighted average for attention. $\star$ indicates significantly worse than best model (bold).
**Sequence-agnostic vs. Sequence-preserving.** Commonly used attention mechanisms simply average the vectors in the focus of attention. This means that sequential information is not preserved. We use the term sequence-agnostic for this. In contrast, we propose to investigate sequence-preserving attention as presented in . We expect this to be important for many NLP tasks. Sequence-preserving attention is similar to k-max pooling which also selects an ordered subset of inputs. While traditional k-max pooling is unweighted, our sequence-preserving ways of attention still make use of the attention weights.
------ ----------- ----------- --------- ---------
all k-max per-dim per-pos
Wiki 67.08 **67.52** 66.73 66.50
Bio **85.57** 84.36 84.05 84.03
------ ----------- ----------- --------- ---------
: $F_1$ results for UD. Model: CNN, S=E, F=W (13). Sequence-agnostic vs. sequence-preserving attention.
compares k-max pooling, attention and two “hybrid” designs, as described in . We run these experiments only on the CNN with external attention focused on word embeddings (, line 13), the best performing configuration in the previous experiments.
First, we investigate what happens if we “discretize” attention and only consider the values with the top k attention weights. This increases performance on wiki (from 67.08 to 67.52) and decreases it on bio (from 85.57 to 84.36). We would not expect large differences since attention values tend to be peaked, so for common values of k ($k \geq 3$ in most prior work on k-max pooling) we are effectively comparing two similar weighted averages, one in which most summands get a weight of 0 (k-max average) and one in which most summands get weights close to 0 (average over all, i.e., standard attention).
Next, we compare sequence-agnostic with sequence-preserving attention. As described in , two variants are considered. In k-max pooling, we select the k largest weighted values per dimension (per-dim in ). In contrast, k-max sequence (per-pos) selects all values of the k positions with the highest attention weights.
In , the sequence-preserving architectures are slightly worse than standard attention (i.e., sequence-agnostic averaging), but not significantly: performance is different by about half a point. This shows that k-max sequence and attention can similarly be used to select a subset of the information available, a parallel that has not been highlighted and investigated in detail before.
Although in this case, sequence-agnostic attention is better than sequence-preserving attention, we would not expect this to be true for all tasks. Our motivation for introducing sequence-preserving attention was that the semantic meaning of a sentence can vary depending on where an uncertainty cue occurs. However, the core of uncertainty detection is keyword and keyphrase detection; so, the overall sentence structure might be less important for this task. For tasks with a stronger natural language understanding component, such as summarization or relation extraction, on the other hand, we expect sequences of weighted vectors to outperform averaged vectors. In Section \[outlook\], we show that sequence-preserving attention indeed improves results on a sentiment analysis dataset.
Comparison to State of the Art
------------------------------
Table \[tab:soa\] compares our models with the state of the art on the uncertainty detection benchmark datasets. On Wikipedia, our CNN outperforms the state of the art by more than three points. On bio, the best model uses a large number of manually designed features and an exhaustive corpus preprocessing [@tang]. Our models achieve comparable results without preprocessing or feature engineering.
Model wiki bio
--------------------------------- ----------- -----------
SVM [@georgescul] 62.01 78.64
HMM [@hmm] 63.97 80.15
CRF + ling [@tang] 55.05 **86.79**
Our CNN with external attention **67.52** 85.57
: Comparison of our best model with the state of the art[]{data-label="tab:soa"}
Analysis
========
Analysis of Attention
---------------------
In an analysis of examples for which pooling alone (i.e., the standard CNN) fails, but attention correctly detects an uncertainty, two patterns emerge.
In the first pattern, we find that there are many cues that have more words than the filter size (which was 3 in our experiments), e.g., “it is widely expected”, “it has also been suggested”. The convolutional layer of the CNN is not able to detect phrases longer than the filter size while for attention there is no such restriction.
The second pattern consists of cues spread over the whole sentence, e.g., “Observations of the photosphere of 47 Ursae Majoris *suggested* that the periodicity *could not* be explained by stellar activity, making the planet interpretation *more likely*” where we have set the uncertainty cues that are distributed throughout the sentence in italics. shows the distribution of external attention weights computed by the CNN for this sentence. The CNN pays the most attention to the three words/phrases “suggested”, “not” and “more likely” that correspond almost perfectly to the true uncertainty cues. K-max pooling of standard CNNs, on the other hand, can only select the k maximum values per dimension, i.e., it can pick at most k uncertainty cues per dimension.
{width=".45\textwidth"}
{width=".4\textwidth"}
#### Pooling vs. Internal vs. External Attention.
Finally, we compare the information that pooling, internal and external attention extract. For pooling, we calculate the relative frequency that a value from an n-gram centered around a specific word is picked. For internal and external attention, we directly plot the attention weights $\alpha_i$. shows the results of the three mechanisms for an exemplary sentence. For a sample of randomly selected sentences, we observed similar patterns: Pooling forwards information from different parts all over the sentence. It has minor peaks at relevant n-grams (e.g. “was sometimes known as” or “so might represent”) but also at non-relevant parts (e.g. “Alternatively” or “the same island”). There is no clear focus on uncertainty cues. Internal attention is more focused on the relevant words. External attention finally has the clearest focus. (See appendix for more examples.)
Analysis of CNN vs RNN {#CNNvsRNN}
----------------------
While the results of the CNN and the RNN are comparable on bio, the CNN clearly outperforms the RNN on wiki. The datasets vary in several aspects, such as average sentence lengths (wiki: 21, bio: 27)[^4], size of vocabularies (wiki: 45.1k, bio: 25.3k), average number of out-of-vocabulary (OOV) words per sentence w.r.t. our word embeddings (wiki: 4.5, bio: 6.5), etc. All of those features can influence model performance, especially because of the different way of sentence processing: While the RNN merges all information into a single vector, the CNN extracts the most important phrases and ignores all the rest. In the following, we analyze the behavior of the two models w.r.t. sentence length and number of OOVs.
shows the $F_1$ scores on Wikipedia of the CNN and the RNN with external attention for different sentence lengths. The lengths have been accumulated, i.e., index 0 on the x-axis includes the scores for all sentences of length $l \in [0,10)$. Most sentences have lengths $l < 50$. In this range, the CNN performs better than the RNN but the difference is small. For longer sentences, however, the CNN clearly outperforms the RNN. This could be one reason for the better overall performance.
{width=".4\textwidth"}
A similar plot for $F_1$ scores depending on the number of OOVs per sentence does not give additional insights into the model behaviors: The CNN performs better than the RNN independent of the number of OOVs (Figure in appendix).
Another important difference between CNN and RNN is the distribution of precision and recall. While on bio, precision and recall are almost equal for both models, the values vary on wiki:
P R
-------------------------- ------ ------
CNN 52.5 85.1
CNN + external attention 58.6 78.3
RNN 75.2 49.6
RNN + external attention 76.3 52.0
Those values suggest that the RNN predicts uncertainty more reluctantly than the CNN.
Outlook: Different Task {#outlook}
=======================
To investigate whether our attention methods are also applicable to other tasks, we evaluate them on the 2-class Stanford Sentiment Treebank (SST-2) dataset[^5] [@sst]. For a baseline model, we train a CNN similar to our uncertainty CNN but with convolutional filters of different widths, as proposed in [@kim2014], and extend it with our attention layer. As cues for external attention, we use the most frequent positive phrases from the train set. Our model is much simpler than the state-of-the-art models for SST-2 but still achieves reasonable results.[^6]
Model S F test set
-------------------- --- --- -----------
Baseline CNN - - 84.84
CNN attention-only I H 83.56
CNN combined I H 85.22
CNN combined I W 86.11
CNN combined E H 86.06
CNN combined E W **86.89**
: Accuracy on SST-2, different focus and source of attention.[]{data-label="tab:sentiment1"}
------- ------- --------- -----------
all k-max per-dim per-pos
86.89 86.39 87.00 **87.22**
------- ------- --------- -----------
: Accuracy on SST-2, sequence-agnostic vs. sequence-preserving attention.[]{data-label="tab:sentiment2"}
The results in Table \[tab:sentiment1\] show the same trends as the CNN results in , suggesting that our methods are applicable to other tasks as well. Table \[tab:sentiment2\] shows that the benefit of sequence-preserving attention is indeed task dependent. For sentiment analysis on SST-2, sequence-preserving methods outperform the sequence-agnostic ones.
Related Work {#relwork}
============
**Uncertainty Detection.** Uncertainty has been extensively studied in linguistics and NLP [@kiparsky1968; @karttunen1973; @karttunen], including modality [@sauri2012; @stanfordFactbank; @szarvas2012] and negation [@velldal2012; @baker2012]. , and conducted cross domain experiments. Domains studied include news [@factbank], biomedicine [@bioscope], Wikipedia [@weasel] and social media [@twitter]. Corpora such as FactBank [@factbank] are annotated in detail with respect to perspective, level of factuality and polarity. conducted uncertainty detection experiments on a version of FactBank extended by crowd sourcing. In this work, we use CoNLL 2010 shared task data [@CoNLLsharedTask] since CoNLL provides larger train/test sets and the CoNLL annotation consists of only two labels (certain/uncertain) instead of various perspectives and degrees of uncertainty. When using uncertainty detection for information extraction tasks like KB population (), it is a reasonable first step to consider only two labels.
**CNNs.** Several studies showed that CNNs can handle diverse sentence classification tasks, including sentiment analysis [@kalchbr; @kim2014], relation classification [@zeng2014; @dosSantos2015] and paraphrase detection [@abcnn]. To our knowledge, we are the first to apply them to uncertainty detection.
**RNNs.** RNNs have mainly been used for sequence labeling or language modeling tasks with one output after each input token [@rnnlm01; @rnnlm02]. Recently, it has been shown that they are also capable of encoding and restoring relevant information from a whole input sequence. This makes them applicable to machine translation [@gatedRNN; @attention0] and sentence classification tasks [@rnnRelClass; @attention2; @attention3]. In this study, we apply them to UD for the first time and compare their results with CNNs.
**Attention** has been mainly used for recurrent neural networks [@attention0; @attention1; @attention2; @attention3; @attention4; @attention5]. We integrate attention into CNNs and show that this is beneficial for uncertainty detection. Few studies in vision integrated attention into CNNs [@vision01; @vision02; @vision03] but this has not been used often in NLP so far. Exceptions are , and . used several layers of local and global attention in a complex machine translation model with a large number of parameters. Our reimplementation of their network performed poorly for uncertainty detection (51.51/66.57 on wiki/bio); we suspect that the reason is that ’s training set was an order of magnitude larger than ours. Our approach makes effective use of a much smaller training set. compared attention based input representations and attention based pooling. Instead, our goal is to keep the convolutional and pooling layers unchanged and combine their strengths with attention. applied a convolutional layer to compute attention weights. In this work, we concentrate on the commonly used feed forward layers for that. Comparing them to other options, such as convolution, is an interesting direction for future work.
Attention in the literature computes a weighted average with internal attention weights. In contrast, we investigate different strategies to incorporate attention information into a neural network. Also, we propose external attention. The underlying intuition is similar to attention for machine translation, which learns alignments between source and target sentences, or attention in question answering, which computes attention weights based on a question and a fact. However, these sources for attention are still internal information of the network (the input or previous output predictions). Instead, we learn weights based on an external source – a lexicon of cue phrases.
Conclusion
==========
In this paper, we presented novel attention architectures for uncertainty detection: external attention and sequence-preserving attention. We conducted an extensive set of experiments with various configurations along different dimensions of attention, including different focuses and sources of attention and sequence-agnostic vs. sequence-preserving attention. For our experiments, we used two benchmark datasets for uncertainty detection and applied recurrent and convolutional neural networks to this task for the first time. Our CNNs with external attention improved state of the art by more than 3.5 $F_1$ points on a Wikipedia benchmark. Finally, we showed in an outlook that our architectures are applicable to sentiment classification as well. Investigations of other sequence classification tasks are future work. We made our code publicly available for future research (<http://cistern.cis.lmu.de>).
Acknowledgments {#acknowledgments .unnumbered}
===============
Heike Adel is a recipient of the Google European Doctoral Fellowship in Natural Language Processing and this research is supported by this fellowship.
This work was also supported by DFG (SCHU2246/8-2).
Supplementary Material
======================
Parameter Tuning
----------------
All parameters and learning rate schedule decisions are based on results on the development set (20% of the official training set). After tuning the hyperparameters (see Tables \[tab:params1\] and \[tab:params2\]), the networks are re-trained on the whole training set.
We trained the CNNs with stochastic gradient descent and a fixed learning rate of 0.03. For the RNNs, we used Adagrad [@adagrad] with an initial learning rate of 0.1. For all models, we used mini-batches of size 10 and applied L2 regularization with a weight of 1e-5. To determine the number of training epochs, we looked for epochs with peak performances on the development set.
-- ------- --------- -------- ----------- --------
Model \# conv filter \# hidden \# att
filters width units hidden
units
(3) 200 3 200 -
(6) 100 3 500 -
(7) 200 3 200 -
(11) 200 3 200 -
(12) 200 3 200 200
(13) 100 3 200 200
(3) 200 3 500 -
(6) 100 3 200 -
(7) 100 3 500 -
(11) 200 3 200 -
(12) 200 3 500 100
(13) 200 3 50 100
-- ------- --------- -------- ----------- --------
: Result of parameter tuning for CNN (“att hidden units” is the number of units in the hidden layer of the attention component); Model numbers refer to numbers in the main paper []{data-label="tab:params1"}
-- ------- -------- ----------- --------
Model \# rnn \# hidden \# att
hidden units hidden
units units
(2) 10 100 -
(4) 10 100 -
(5) 10 200 -
(8) 10 100 -
(9) 30 200 200
(10) 10 200 100
(2) 10 500 -
(4) 10 500 -
(5) 10 50 -
(8) 10 50 -
(9) 30 100 200
(10) 10 50 200
-- ------- -------- ----------- --------
: Result of parameter tuning for RNN[]{data-label="tab:params2"}
Additional Examples: Attention Weights
--------------------------------------
{width=".42\textwidth"}
{width=".38\textwidth"}
and compare pooling, internal attention and external attention for randomly picked examples from the test set. Again, pooling extracts values from all over the sentence while internal and external attention learn to focus on words which can indicate uncertainty (e.g. “thought” or “probably”).
Additional Figure for Analysis: Results Depending on Number of OOVs
-------------------------------------------------------------------
plots the $F_1$ scores of the CNN and RNN with external attention w.r.t. the number of out-of-vocabulary (OOV) words in the sentences. The number of OOVs have been accumulated, i.e., index 0 on the x-axis includes the score for all sentences with a number of OOVs in \[0,10), etc.
{width=".45\textwidth"}
[^1]: <http://cistern.cis.lmu.de>
[^2]: Gates, e.g., the weighting of $h^{t-1}$ in , can also be viewed as internal attention mechanisms.
[^3]: The small difference between the RNN results on bio on lines (5) and (8) is not significant.
[^4]: number of tokens per sentence after tokenization with Stanford tokenizer [@stanfordTok].
[^5]: <http://nlp.stanford.edu/sentiment>
[^6]: The state-of-the-art accuracy is about 89.5 [@sst-soa; @mvcnn].
| null |
minipile
|
NaturalLanguage
|
mit
| null |
I stopped thinking people who wrote about things like free speech read the rudimentary texts through which the roots of those rights run a long time ago. Given some of the absurd attacks on the near-universality of the right to free speech in light of the use of an anti-Islamic video that ignited anti-American protests around the Muslim world as a stand-in by the U.S. government for any of its own actual policies or actions that might have fueled and sustained such protests as seen over the last two weeks, here is a relevant passage on why the right to express an opinion is a much more precious (not to mention real!) one than the right not to be offended, from Jon Stuart Mill's "On Liberty":
The time, it is to be hoped, is gone by when any defence would be necessary of the "liberty of the press" as one of the securities against corrupt or tyrannical government. No argument, we may suppose, can now be needed, against permitting a legislature or an executive, not identified in interest with the people, to prescribe opinions to them, and determine what doctrines or what arguments they shall be allowed to hear. This aspect of the question, besides, has been so often and so triumphantly enforced by preceding writers, that it needs not be specially insisted on in this place. Though the law of England, on the subject of the press, is as servile to this day as it was in the time of the Tudors, there is little danger of its being actually put in force against political discussion, except during some temporary panic, when fear of insurrection drives ministers and judges from their propriety; and, speaking generally, it is not, in constitutional countries, to be apprehended that the government, whether completely responsible to the people or not, will often attempt to control the expression of opinion, except when in doing so it makes itself the organ of the general intolerance of the public.
Let us suppose, therefore, that the government is entirely at one with the people, and never thinks of exerting any power of coercion unless in agreement with what it conceives to be their voice. But I deny the right of the people to exercise such coercion, either by themselves or by their government. The power itself is illegitimate. The best government has no more title to it than the worst. It is as noxious, or more noxious, when exerted in accordance with public opinion, than when in opposition to it. If all mankind minus one, were of one opinion, and only one person were of the contrary opinion, mankind would be no more justified in silencing that one person, than he, if he had the power, would be justified in silencing mankind. Were an opinion a personal possession of no value except to the owner; if to be obstructed in the enjoyment of it were simply a private injury, it would make some difference whether the injury was inflicted only on a few persons or on many. But the peculiar evil of silencing the expression of an opinion is, that it is robbing the human race; posterity as well as the existing generation; those who dissent from the opinion, still more than those who hold it. If the opinion is right, they are deprived of the opportunity of exchanging error for truth: if wrong, they lose, what is almost as great a benefit, the clearer perception and livelier impression of truth, produced by its collision with error.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Q:
Average by groups in a sequence of 3D images
following situation. I have a time-series of 3-Dimensional images in a 4D matrix. The resolution of the 3D images is 177 * 209 * 156.
Then I have a 3D-Mask image, which carries the information to which one of 500 groups a voxel in the 3D-Image belongs, so it has the same dimension as one 3D image, but the values are [2;501] (502 actually, but the last group is "outsiders" and those will be ignored).
Now for those 3D images, I want a 500 * X matrix, which holds the average value for each group for each Image, where X is the image-count (which is 1500 in this case). The code I wrote to generate this vector (for each image) is the following:
...
final_vec = zeros(500, TRs);
mask_hist = get_histogram(mask_mat);
mask_hist(1) = [];
mask_hist(end) = [];
for i = 1:TRs
TR = new_nifti_mat(:,:,:,i);
for k = 1:numel(TR)
if mask_mat(k) < 502 && mask_mat(k) > 1
final_vec(mask_mat(k) - 1, i) = final_vec(mask_mat(k) - 1, i) + TR(k);
end
end
final_vec(:, i) = final_vec(:, i)./mask_hist
end
...
The Problem with this code is, that it runs for ever. For one set of data it ran more than 12h. I know that matlab is bad with for-loops, and probably has a very elegant 1 or 2 line of codes way to do this in a fraction of the time it needs for my code
Best Regards
Uzaku
A:
Let's create example data first:
x = rand(4,4,2,8); %// random data
mask(:,:,1) = [ 3 2 2 3
2 4 4 2
2 4 4 2
3 2 2 3 ];
mask(:,:,2) = mask(:,:,1); %// example 4x4x2 mask
Approach 1: one loop and accumarray
Loop over each 3D images, and for each one use accumarray to do the average by groups:
mask = mask - min(mask(:)) + 1; %// so that lowest value is 1
result = NaN(max(mask(:)),size(x,4)); %// preallocate result
for n = 1:size(x,4);
result(:,n) = accumarray(mask(:), reshape(x(:,:,:,n),[],1), [], @mean, NaN);
end
Approach 2: bsxfun and matrix multiplication; no loops
Linearize the first three dimensions and build a zero-one mask so that accumulation of all values within each group becomes matrix multiplication:
mask2 = bsxfun(@eq, mask(:), min(mask(:)):max(mask(:))).'; %'
result = mask2*reshape(x,[],size(x,4)); %// sum for each group
result = bsxfun(@rdivide, result, sum(mask2,2)); %// transform sum into average
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Low Cost Dumpster Rentals
Portable Dumpster in Barstow, TX
Free Price Quote, Call Now!
Need a Dumpster Rental in Barstow?
We have the commercial and residential dumpster service you need in Barstow.
At Dumpster Centre, we strive to bring you the best service at the best price available. We have a wide selection of dumpster sizes in stock in Barstow, TX, with flexible & timley pickup and delivery. Give us a call today for all your dumpster rental and roll off needs.
Give us a call today for a free dumpster quote. (432) 242-6728
More About Us
Dumpster Centre has made it super fast and easy to rent a dumpster. You only need 5 minutes!
When you are renting a 5, 10, 15, 20, 30 or 40 yard dumpster, you want a company you can trust with prices that make you smile. Give us a call today and see the difference we can make in your next construction or clean out project.
Free Consultation
Simply give us a call and we will help you figure out your dumpster rental needs.
Prompt Delivery
Our dumpsters usually go out same-day or next-day depending on when you call.
Competitive Pricing
We provide top-notch service, while going easy on your bottom line. What more could you ask for?
Hassle-Free Ordering
Our trained operators are here to give you a fast and hassle-free experience from start to finish.
Fast Pickup
Once you are done, simply give us a call and we will get your dumpster picked up right away!
Responsive Staff
Our staff is here to help whenever you need it. Simply pick up the phone and give us a call.
Get a Dumpster Quote Now!
Operators are Standing By
Our Dumpster Specialists are waiting by the phone to help you get started. However, there is some important information we need before we can give you a price quote and delivery time. Some things we need to know:
What size dumpster you will need
What you plan on putting in your dumpster
When you need us to drop your roll off container off
How long you plan on keeping your roll-off
City & zip code where you plan on placing your dumpster
Ready to get started?
Dumpsters are available* in Barstow.
Simply give us a call now at (432) 242-6728 to get started! We will help you pick a dumpster for your needs, schedule your delivery and take payment… all in less than 10 minutes!
Zip Codes in Service Area
Free Price Quote, Call Now!
Dumpster Rental Info Center
City Regulations on Dumpster Rental
Most cities or municipalities do not have many regulations regarding dumpster rental as long as you keep the dumpster entirely on your own property during the rental period. If you need to place your container on the street in any way, you will likely have to obtain a permit from the appropriate building permit office in your town. You can check with your local public works department to be sure.
Most Barstow dumpster rental companies will take care of securing this permit for you if you rent from them. Make sure that if you’re planning to put the dumpster on the street, the company has made the appropriate arrangements. You should also ensure that you get the permit in a timely manner and at the correct cost. If you thought the dumpster company was getting a permit and they did not, you will be the one who will have to pay the fine that is issued by the authorities.
One Clean Sweep Home Construction Dumpster Rental in Barstow, TX
When phoning a service you will want to speak someone as opposed to a piece of equipment recording, as there are a number of different styles and sizes of dumpsters any particular one would most likely want to some help choosing which package will best fit their needs. Some dimensions and canisters are better in shape for certain homes; you would not rent the most significant size regarding adding patio decking onto your property, nor could you buy the tiniest dumpster for house building demolition. The roll off of container is obviously not a thing that would be bought either, so when it comes to rubbish removal, in which the options grow to be increasingly constrained, a dumpster is the ideal tool for the task. After the right size of your container is decided, the different products being hauled will be decided, since hazardous products and very fragile things such as cup are at a much higher risk connected with causing deterioration due to improper handling, and as a consequence are taken care of differently. If there's a possibility connected with recycling products, that will be discovered also, seeing that companies are and really should well be, becoming more ecologically conscious.
When compared to letting a dumpster, buying a Big Carrier is an excellent choice. A Big Carrier has a big volume--it holds around 6 cubic feet connected with rubbish, which can be comparable to a normal dumpster. Unlike any roll-off dumpster, which requirements to be located so it can be unloaded and also reloaded onto any truck, the fundamental Bag might be set up anywhere you need it to possibly be. You don't need any permit of any sort to use a Massive Bag, so you won't threat any deterioration to your property from your bag itself, whereas any dumpster can split up your yard or gouge the driveway relatively easily. Moreover to these positive aspects, buying and utilizing a Big Carrier can be a ton cheaper than Portable Dumpster in Barstow, TX. Luggage cost a fair amount, usually in the $30 variety. With a case, you decide the pick-up date for all of your debris, which means you don't have to prepare yourself at the end of the Portable Dumpster in Barstow, TX period or perhaps be liable for late payment fees if you pass up your deadline day. And, as the bag can be yours to maintain, you simply currently have to pay for your service to pick-up your next heap whenever you want to apply it again. Some companies will even provide discounts on their hauling rates if you use the bag oftener. When it's not being used, you can simply collapsible your case: It takes upwards about as much space as a large backpack when it's stored.
There are some deterring measures that your user may miss out while dealing with your waste made if it is taken care of on home. Initially the consumer will be unaware of the size of your dumpster that is required to clear the trash at one head out. Picking an unacceptable sized dumpster may lead to doubling the work as well as cost in terms of carry. There exists varying number of rules from city to city which the user offers to know thyself to get rid of the trash.
Benefits of Rolloff Dumpsters in Barstow
What is the easiest way in order to recycle the garbage or home the trash in the landfill? In case you have decided to home your trash without damaging environment, Portable Dumpster in Barstow could be the easiest way to accomplish it. Recycling the garbage will not only abate the effects connected with pollution within the environment, yet convert the garbage into re-usable merchandise. There are number of Portable Dumpster in Barstow service firms in the market to help you to handle your own waste supervision needs efficiently. They sell or home trash to your landfill in your stead. By placing an order with the Portable Dumpster in Barstow firms, you can have the dumpster placed for your desired spot at the time specified by you. You can fill the dumpster with the trash enjoying your own free time plus the trash will probably be disposed to your landfill upon completion of assembling your shed. These Portable Dumpster in Barstow firms have various sizes of dumpster that best suits your own customized needs. If you are not mindful of right dumpster sizing that fits as part of your requirement, you may also get some help from these Portable Dumpster in Barstow firms and get the dumpster rented on very affordable price.
A specialty support that requires Portable Dumpster in Barstow can be community cleaning programs along with charity sort projects. Usually dumpster company works with the organizer of such associated with an event and frequently times implement discounts for the sake of the community. The service such as this that makes the professional trash solution firms stand out from the remaining and make these individuals the one quit company that one can go to for his or her dumpster needs.
Someone that is performing a home renovation venture would most probably need a key place to store all of the trash and trash they will build up. During makeovers, there are several components such as porcelain tile, wallpaper, kitchen cabinetry and kitchen counter tops, a few. Most people will not have or have immediate access to a trash bin just right to store the many material.
Your current question right here would be the reason why I should choose professional support rental provider once i can gather things and also dump this for home. Let me tell you is capable of doing self throw away management things to do when the trash is tiny but what about your throw away that is accumulated is more? You'd defiantly should create an area to gather your throw away that is where by dumpster is required.
If you are planning on doing renovating compared to you will need to rent any dumpster. By renting any dumpster you can easily maintain the condo spot looking well put together. You do not want the neighbors worrying at all of the mess available nor would you like them to report you to the authorities or your local building home inspector.
through your regional junk or trash haulers gives a number of positive aspects that can make your health much simpler when you're tackling residence projects. If you rent a dumpster, you can get rid of anything, large or small, you need to, you can work on your own schedule, and you can abandon the final cleaning work to the garbage and dumpster advantages.
Using a group construction dumpster help make trash discretion much easier and also efficient since it does not require trash to be quit in your community. Keep the community clean and safe together with the community structure Portable Dumpster in Barstow. Your community looks its greatest while the group works together while using community structure Portable Dumpster in Barstow.
How to Make your like simplier With Roll-Off Dumpster Rental Through your Local Junk Removal Assistance in Barstow, TX
As an productive member in your neighborhood, as a income tax payer, and a U.S. person, we all take extreme take great pride in in the repair of our group. Waste dumpsters promote cleaner local communities as well as neighborhoods, and cleaner life styles. Dumpsters are usually easy and hassle-free ways to discard waste materials accumulating in a household, work place, institution, church, and also places intended for commercial use. Now inside your there is recognition to the effects of improperly disposed waste materials on the globe. Being aware of these types of effects along with knowing how to lessen the problems being done to our planet is the best approach to better our universe. Construction dumpster leases are not confined to construction internet sites but far better neighborhoods everywhere you go.
Are you any resident with the great Atl city? Have you been worried about taking away waste products along with filths around your own home and environments? You don't need to fear any longer! Atl Roll off Dumpsters are now offered at your beck along with call. In fact, waste products can certainly constitute an annoyance when they are not properly removed.
Like a builder you are aware that a lot of items are needed whenever building a household. Most of the items come with plastic material wrapping, Styrofoam, cardboard containers, etc. You would like to continue with constructing the home and getting rid of the particular garbage accruing since the building challenge moves combined.
Before you decide about the Portable Dumpster in Barstow, TXs for the business or perhaps residential use, it is important that you recognize the facts with regards to these leases so you can make a wise decision. Retain these facts planned as you are with the right selection for your spend management needs. You should also be certain to compare organizations, what they have to supply, and their charges before making your responsibility. It's important to find a quality business that provides high quality customer service along with reasonable prices that will fit into your finances.
Now that you understand which following day Portable Dumpster in Barstow, TX to choose for your current construction employment, you are ready to rent that and get your projects underway. It's nice to understand you can save cash when you don't order any next day Portable Dumpster in Barstow, TX that may be too large for the specific development job.
When you do anything, you may need to go into the home and look at the situation. Would the actuel leave private property at the rear of? Trash? What are the damage does the actuel do to the exact property itself? Could it be necessary to substitute the floor coverings, patch holes in the wall structure, or paint? Perform a comprehensive examination of the exact property to figure out what specifically will have to be exercised to make it possible for brand-new tenants to dwell there in the foreseeable future.
Once you have picked an appropriate receptacle based on your family needs the company you ultimately choose will produce it to your property. You have to make a decision about the best location for the particular receptacle. Should the receptacle is necessary in your home, a good option is the garage. Most of the receptacles come with tires and this makes it simple to move all of them.
Request a Call Back!
Still need a quote but don't want to talk on the phone right now? Fill out the form below and we will get back to you!
Please use the form below to contact us. We look forward to working with you and will get back to you as soon as possible. All fields are required
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Alberta Carbon Trunk Line
The Alberta Carbon Trunk Line (ACTL) is a pipeline in its construction stages that will collect excess carbon dioxide (CO2) from the province of Alberta and transport it to various oil reservoirs around the province for enhanced oil recovery applications. Pioneered by Enhance Energy , the ACTL will source its CO2 from a fertilizer manufacturing plant in Redwater, and the North West Redwater Partnership bitumen upgrading plant using gasification. It will initially funnel the collected CO2 to Clive, a field discovered in the 1950s and currently produces about 300 barrels per day.
Expected specifications of captured CO2
The ACTL is expected to be the world's largest carbon capture and storage project. It will store of CO2 per year, six times more than the Weyburn project in Saskatchewan. Initially, the project should collect and transport of CO2 per day and is expected to expand to of CO2 per day.
Timeline
Plans for the ACTL began in 2004, but were put on hold due to the Great Recession in 2008. However, the ACTL was restarted after the recession with a license for construction and operation issued in April 2011. The ACTL was originally planned to start production in 2013, and delayed to 2015, and then finally delayed further. The pipeline is currently in its final stage of production as of November 2016. The ACTL has planned for CO2 injection to begin in early 2018.
Economics
The project has been estimated to cost CAN$1.2 billion but will find approximately CAN$558 million from several energy funds and carbon capture technology initiatives. Over 15 years, the Alberta province will be funding the project with CAN$495 million from the Alberta CCS Fund. The Canadian government is giving a total of CAN$63 million: CAN$30 million from the Clean Energy Fund and CAN$33 million from ecoENERGY Technology Initiative.
References
Category:Buildings and structures in Alberta
Category:Petroleum production
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Working Hard or Hardly Working?
A Facebook status update here, a Tweet there and maybe you find that perfect dinner recipe on Pinterest. In this day and age most jobs require a computer, most computers have the Internet and distractions to workers are plentiful. But how much time do employees waste on visiting personal websites that aren't work-related during the day? And on a more philosophical note, should that really be considered a "waste" of time? We surveyed more than 3,200 people from February to March 2012 to find out.
Blame It on the Internet
As you can probably imagine in this era of constant connectivity, the Internet is forever at our disposal via laptops, smartphones and tablets. Ideally, that's fantastic news for people who want to take their work with them wherever they go. But the flip side is the constant temptation to slack off. Of the people we surveyed, 64 percent said they visit non-work related websites every day during work hours. However, that number is down nearly 10 percent from the last time we conducted this survey in 2008. With so many jobs lost in the last four years, it's likely employees have less time to waste because they're spending more time on their added job responsibilities.
How Much Time is Too Much?
With almost two-thirds of all respondents reporting they waste time at work on the computer each day, the next obvious question is how much time? Thirty-nine percent of the people who took our survey said they spend a mere 1 hour a week or less on non-work related items. That’s followed by 29 percent who spend up to 2 hours a week wasting time on the computer at work, and 21 percent who waste up to 5 hours a week. Only 3 percent of respondents spend 10 hours or more on personal tasks while at work in a given week.
What Websites are Employees Visiting?
We asked our 3,200 respondents what websites they visit if they do stray from work-related tasks. Most people spend time checking their personal email, visiting news sites, performing Google searches, monitoring social media and shopping online. It probably comes as no surprise Facebook topped the list. The social media behemoth with nearly 850 million users worldwide is visited by 41 percent of our respondents. That’s followed closely by LinkedIn at 37 percent, Yahoo at 31 percent, Google+ at 28 percent and Amazon.com with 25 percent. Twitter ranked near the bottom at a mere 8 percent. And even though Pinterest has been garnering a lot of media attention lately, only 4 percent of our respondents currently use the service.
Why Waste Time?
We're Salary.com, which means we tend to focus a lot on compensation. So if employees are wasting time instead of working, it stands to reason they might be doing it because they're underpaid, right? Wrong. Of the top six reasons why employees waste time at work, being underpaid ranked dead last at 18 percent. Most employees -- 35 percent -- said they waste time at work because they're not challenged enough. That was followed closely by the 34 percent of employees who claimed they waste time because their hours are too long, 32 percent whose company gives them no incentive to work harder, and 30 percent who are unsatisfied. Additionally, 23 percent of respondents said they waste time at work simply because they're bored.
Breaking Down the Numbers
Men waste more time than women at work. More than two-thirds -- 69 percent -- of men reported using the Internet for personal reasons during work hours on a daily basis, compared to 62 percent of women. And even though it's assumed by many that young people spend more time on websites that aren't work related during the workday, that's not the case. Workers between the ages of 26-35 topped the list with 75 percent wasting time at work on a daily basis, compared to the 72 percent of 18- to 25-year-olds. Daily Internet use at work dropped off among older workers with 65 percent of people age 36-45, 58 percent of people age 46-55 and 55 percent of people 56 and older wasting time on the Internet while at work every day. Even among those who waste more than 10 hours a week at work, the 18-25 group comes in third (15 percent) behind workers 26-35 (35 percent) and 36-45 (29 percent).
Does Education = Wasteful?
Although it's never a waste of time to get a good education, our survey shows the more educated you are the more time you waste at work. Only 59 percent of high school graduates reported wasting time at work on a daily basis. Compare that to the 67 percent of respondents with doctorate degrees. Those with PhDs are followed closely by people with bachelor's degrees at 66 percent, and those who have earned master's degrees at 65 percent. One reason for the disparity could be that people with college degrees and higher levels of education are generally promoted to managerial and supervisory roles.
Restricting Access Might Backfire
Thirty percent of respondents said personal sites like Facebook and Twitter are blocked at work. Unfortunately for the companies that block personal websites, their efforts might be fruitless because 60 percent of the people surveyed said they would simply use their own smartphones, tablets and laptops to access personal websites during work hours. A recent report by KPMG International addressed the ramifications of employers choosing to block personal websites at work. “Executives may be naïve in thinking that banned access to social networks eliminates employee use,” suggested Tudor Aw, KPMG’s European head of Technology. “Indeed, the survey shows that by restricting or blocking access, many employees tend to move their activity to their own personal devices which are often less secure and completely unmonitored.”
No Access? No Big Deal
With social media ingrained into the everyday lives of most workers, how do employees feel about not being able to access their personal sites? Apparently it's not a big deal. A whopping 79 percent of people who took our survey said they simply don't care whether or not employers block certain websites, and that it would not affect their decision to work there. And while 18 percent said the restrictions would make them look unfavorably on the company in question, 4 percent said the company's no-nonsense approach actually makes it a more attractive place to work.
Job-Hunting on the Job
Although 70 percent of respondents said their employers don't put any restrictions for online use in place, maybe companies do have some valid fears. Businesses pay millions of dollars every year to retain employees and cut down on turnover costs. Therefore those companies might have a problem with the fact that nearly half -- 46 percent -- of those surveyed said they've spent time job hunting during work hours and on company computers.
Wasted Time Occurs Offline Too
Companies that solely blame the Internet for wasted time aren't seeing the big picture, and might benefit from a look in the mirror. Only 18 percent of those surveyed listed the Internet as a time-waster. The biggest waste of time, according to 47 percent of our respondents, is having to attend too many meetings. That's followed by dealing with office politics (43 percent), fixing other peoples' mistakes (37 percent), coping with annoying coworkers (36 percent), busy work (22 percent) and returning an abundance of work emails (20 percent). Dealing with bosses came in last at a mere 14 percent.
Is "Waste" the Wrong Word?
With all this talk of "wasting time" during work by checking Facebook, chatting with coworkers, etc., it's a worthwhile question to ask: are those things truly time-wasters? Not according to the people who took our survey. Nearly three-quarters -- 71 percent -- of respondents said they believe short breaks throughout the day are beneficial. By being able to check Facebook, Twitter and have periods of brief downtime throughout the workday, those surveyed said they believe employees will actually be more productive than if restrictions are placed on them.
More Employers Recognizing Cultural Shift
Rudy Karsan, CEO of Kenexa (Salary.com's parent company), was a guest on Salary Talk last year and spoke on this very topic. "Basically, younger workers are coming in and saying 'I'm going to be at my desk, I'll continue working, then I'm going to get distracted by doing some shopping or watching YouTube, then I'll come back and do my job.' Back and forth, in and out. When we start to do that we're really blending our lives together with work. I applaud it and hope we never lose that, because that's natural for us." Karsan, along with many other employers, recognizes the cultural shift taking place and sees the value in creating a "Work-Life Blend" as opposed to a balance. More employees than ever are answering work emails via smartphone well after they've left the office, so a little downtime during the workday becomes no big deal. Which means next year we might have to change the title of this survey!
Recommended Reading
| null |
minipile
|
NaturalLanguage
|
mit
| null |
The loadcell live idler platform scale relates generally to weighing devices for bulk product. More specifically, the present invention refers to conveyor belt scales of the weighbridge type, used for weighing material as it is transported along a conveyor belt. A unique aspect of the present invention is a calibration weight that temporarily loads an idler with a known weight for calibrating the load cells.
For decades, mines, quarries, foundries, mills, and the like have moved bulk product by conveyor from one location to another. A conveyor also moves product between one stage in a process to another. Each stage in a process likely calls for the weight of product entering the process. Conveyor belt scales are used for continuous weighing of bulk materials transported along the conveyor belt. Examples of such materials are coal, aggregate, slag, or wood chips among others. Knowing the weight allows for adjusting a process and results in an end product within tolerances.
In general, conveyor belt scales employ rollers or idlers that are inserted into the conveyor system, either in place of an existing idler assembly or in addition to the existing idlers. The conveyor belt rests on the idlers and the material, coal for example, passes over the idler assembly causing a downward deflection of the system. Generally, the idler assembly is cooperatively connected to a weight sensor. The weight sensor, depending upon the construction of the particular weighing device, converts the downward deflection into a weight measurement. Ideally, a weighing device would combine the weight with a determination of the speed of the moving conveyor belt to yield a rate of flow of material as well as the total weight of the passing material.
The current invention provides an idler having weight measuring load cells and optionally a speed sensor, controller, and a junction box for multiple units. The present invention displays to an operator an accurate reading of the amount of product crossing the invention within a given time interval. Further, the present invention calibrates the weight measurement upon command using a weight connected beneath the idler.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
JadR*-mediated feed-forward regulation of cofactor supply in jadomycin biosynthesis.
Jadomycin production is under complex regulation in Streptomyces venezuelae. Here, another cluster-situated regulator, JadR*, was shown to negatively regulate jadomycin biosynthesis by binding to four upstream regions of jadY, jadR1, jadI and jadE in jad gene cluster respectively. The transcriptional levels of four target genes of JadR* increased significantly in ΔjadR*, confirming that these genes were directly repressed by JadR*. Jadomycin B (JdB) and its biosynthetic intermediates 2,3-dehydro-UWM6 (DHU), dehydrorabelomycin (DHR) and jadomycin A (JdA) modulated the DNA-binding activities of JadR* on the jadY promoter, with DHR giving the strongest dissociation effects. Direct interactions between JadR* and these ligands were further demonstrated by surface plasmon resonance, which showed that DHR has the highest affinity for JadR*. However, only DHU and DHR could induce the expression of jadY and jadR* in vivo. JadY is the FMN/FAD reductase supplying cofactors FMNH₂/FADH₂ for JadG, an oxygenase, that catalyses the conversion of DHR to JdA. Therefore, our results revealed that JadR* and early pathway intermediates, particularly DHR, regulate cofactor supply by a convincing case of a feed-forward mechanism. Such delicate regulation of expression of jadY could ensure a timely supply of cofactors FMNH₂/FADH₂ for jadomycin biosynthesis, and avoid unnecessary consumption of NAD(P)H.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Analysis of intrahepatic lymphocyte subsets in a transgenic mouse model of immune-mediated hepatocarcinogenesis.
Long-term, persistent liver cell injury increases the risk for hepatocellular carcinoma (HCC) development in chronic viral hepatitis. In support of this notion, we have developed a unique animal model of chronic immune-mediated liver disease that induces hepatocellular carcinogenesis using HBV transgenic mice; however, the intrahepatic inflammatory response was not precisely evaluated. The current study demonstrated that hepatitis B surface antigen (HBsAg)-specific cytotoxic T lymphocytes (CTLs) were detected at a frequency of 0.05% of CD8+ T lymphocytes in the liver, and that monocytes/macrophages were remarkably increased as the disease developed. These results suggest that a minimal number of intrahepatic virus-specific CTLs and the recruited monocytes/macrophages may contribute to the process of chronic liver inflammation.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
May 5, 2010
Arne opened the meeting with a welcome to all followed by updates and announcements.
Updates and Announcements:
Arne showed a PowerPoint recapping and depicting several events held and/or sponsored by the Library (or the FOSL) during the FY09-10.
What is currently going on at Steely?
Archives - we are in the process of installing compact shelving in anticipation of a large collection. At the end of May (thru August?) room SL102 will be used to temporarily shelve materials from the Archives while the new flooring and shelving is installed.
The Bachelor of Library Informatics (a fully-online) program is doing very well. We are in our second semester and anticipate 50 enrolled for Fall. It is drawing a much larger audience than we first thought. Librarians generally start out in other areas and then become librarians but if we can target interest from high school forward, we can attract more qualified individuals to the profession and they will be better equipped and prepared to be more effective librarians. The program is practitioner taught program, with courses taught by Steely faculty/librarians.
IMLS Grant Program ( NKU, Bluegrass and KDLA) - There are now 50 recipients of scholars and stipends, as well as 12 receiving scholarships only. These students are located throughout state with most students currently attending Bluegrass. KYlibrarians.org, an integrated website for all library education in Kentucky, a statewide mentoring program, and a series of continuing colloquia are also now in operation. We held our first Advisory Board Meeting and received very positive feedback from members of the board.
Budget Issues - The bad news, the Library will receive a $39,000 cut this year. We have restructured our serials and now purchase much of the collection in packages rather than as separate titles. This has worked to our advantage by allowing us to provide much more content even with ongoing cuts, however, we will lose titles in groups rather than singly if we have to cut deeply over time. Serials inflation continues to be a problem and we did not receive coverage this year. Fortunately, we did recoup funds from EOY this year that we will carry forward to address next year’s shortfall. The upside is that we preserve serials; the downside is that we continue to de-emphasize books. Lois Schultz reported that we we recouped $6,000 on Dec. 15th and we’re going to have almost $10,000 with the second cut-off. Nine departments did not use all their funds. If their remaining balance was $150 or less, we counted it as all spent. Vendors are wanting to make “deals”... buy one, get two, etc. but we have been very conservative ordering new items. Reference cancelled some titles and moved some things to full text.
Lois Schultz:
Showed PowerPoint of stats on journal packages which indicated a large usage, in addition, Lois noted what was saved on SourceFinder costs.
Update on Budget Formula - The group met and decided to leave the current formula in place for 1 more year to gain more stats on how it is working - still have the 30% cap. Per Arne, allocations will be very similar to last year.
We are considering placing additional funds into SourceFinder to grow that method of providing content to our users and to build the collection.
New Business:
Katherine Kurk initiated conversation re: the LRC (how it is now vs. when it was in BEP). Since it was relocated to the library, it seems to be getting much less use. Our LRC is one of the nicest in the state, with a fantastic collection, but Katherine feels that it may not being used enough due to the lack of a “production room”. Arne indicted that he had talked with Jenny Smith about it in the past but that we have had space issues in this building. However, there may be some space coming available. Arne also stated that the auditors made us put the lamination equipment behind the circulation desk. All agreed that we are missing out on a big opportunity where teachers/students/grads can learn from one another. Arne said that a room has been planned (when the budget permits) that is glass enclosed, etc. and will be VERY nice. Researching opportunities for private funding.
Arne asked for feedback re: the group, in general, its purpose/usefulness, the # of meetings, etc. Everyone agreed the group served a useful purpose and that quarterly meetings were sufficient, with email/correspondence notifying of any immediate issues.
KC Russell asked about the status of the RFID tags - Arne announced it was going well but we would need additional money in order to continue. This was included in this year’s requests for funding. We hope to be able to complete this project within the next 2 years. Once everything is switched to RFID, we can purchase a new security system to replace the old equipment. All new books are being tagged.
Vicki Culbreth-Berling asked what technology was available for remote students and how was it working? Arne indicated we currently have instant and text-messaging and online chat. The next great shift toward the virtual library is in services. As serials have gradually moved from majority print to majority access online, we should see the same happen with services. We have been focusing on Info Literacy. We now have many online tutorials, for example. Any stats on usage? Are students catching up with technology? Yes and no. We have embedded librarians (approx. 86 courses) which has been very successful and we are working with instructors to help relay research skills. Non-library students are enrolling in the BLI program just to acquire research/resourcing skills to apply to all their classes.
Tom Heard asked, “What are the librarians’ opinions of E-Content”? Arne said we have added more equipment (i.e. printers, scanner, imaging, audio, etc.) and that the general feedback is “good”.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Smooth muscle guanine nucleotides and receptor-effector coupling following inhibition of oxidative energy production.
We measured total concentrations of guanosine triphosphate (GTP) and guanosine diphosphate (GDP) in rabbit aortic smooth muscle under several different conditions. We computed free [GDP] and free GTP/GDP (using a Keq of 1.0 for the nucleoside diphosphate kinase reduction) under these conditions. At a time when muscle was contracted by 15 microM norepinephrine (NE) for 5 min under normoxia, [GTP], free [GDP], and free GTP/GDP were 0.29 +/- 0.03 mM, 3.5 microM, and 82, respectively. Following rapid inhibition of oxidative energy production during NE-evoked maintained force, which is associated with slow decreases in mean tissue [PCr], and [ATP] and force relaxation, [GTP] and free GTP/GDP were decreased at relaxation threshold to 0.22 +/- 0.02 (SE) mM and 43, respectively, and progressively fell further, paralleling decreases in force and [ATP] and [PCr]. There were marked decreases in the sum of GTP + GDP contents under conditions where muscle energy stores were decreased (i.e., low [PCr] + [ATP]). Similar data were obtained during a 50 mM KCl-evoked contracture. Free [GDP] increased from normoxic values of 3.5 microM to values as great as 6.0 microM at low energy store states. Free GDP was equivalent to 6% of total GDP under normoxia and increased to 16-21% of total GDP under conditions of low energy stores. Evidence was obtained that decreases in [GTP] or free GTP/GDP seen under conditions of low total energy stores were not sufficient to inhibit heterotrimeric G protein function and uncouple receptor-effector coupling.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Bilateral Triple Talocalcaneal, Calcaneonavicular, and Talonavicular Tarsal Coalition: A Case Report.
We report a rare case of a female, aged 42 years, with symptomatic bilateral triple tarsal coalition, that is, talocalcaneal, calcaneonavicular, and talonavicular tarsal coalition. The patient was treated conservatively by adjusting her activities. At the 12-month follow-up, the patient was asymptomatic. Bilateral triple tarsal coalition is a rare disorder, especially in nonsyndromic patients. The purpose of this case report was to highlight this rare type of multiple bilateral tarsal coalitions and to discuss the relevant existing literature.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
British Battlegroup
The British Battlegroup or UK Battlegroup is an EU Battlegroup, consisting entirely of British armed forces. It was on standby from 1 January, until 30 June 2005, and from 1 July, until 31 December 2008.
Background
The United Kingdom, as part of the European Union, contributes part of its armed forces to the EU Battlegroup scheme. This scheme puts a battlegroup composed of elements from one or more EU member states at the disposal of the European Council to be deployed as required. As one of the larger EU members, the United Kingdom contributes to two individual battlegroups. The British Battlegroup is made up entirely of British forces, while the other, the UK–Dutch Battlegroup, has the UK as the "framework" (i.e. lead) nation in the formation, in conjunction with forces from the Netherlands. The Netherlands also contributes to Battlegroup 107, alongside Germany and Finland.
Composition and equipment
The two UK-led battlegroups are formed from units declared to the UK's Joint Rapid Reaction Force, a rapid deployment element formed either from 3 Commando Brigade or 16 Air Assault Brigade, which are the elite commando and paratroop formations.
References
Category:Battlegroups of the European Union
Category:Military of the United Kingdom
Category:Military of the Netherlands
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Q:
Calling Elasticsearch Template Query using NEST?
Is there a way to call Template Query using NEST? Is there any examples?
A:
The search template endpoint isn't mapped in NEST yet, and poses a bit of a challenge since it's very different to how queries are normally constructed. We're actually working on this now (in this branch) and are hoping to get this functionality in the upcoming 1.1 release. Here's a link to the original issue for tracking purposes.
EDIT: Forgot to mention, the endpoint is available on the low-level Elasticsearch.Net client, which you can access via ElasticClient:
var client = new ElasticClient(...);
client.Raw.SearchTemplate(...);
| null |
minipile
|
NaturalLanguage
|
mit
| null |
A SequenceSpace analysis of Lys49 phopholipases A2: clues towards identification of residues involved in a novel mechanism of membrane damage and in myotoxicity.
'SequenceSpace' analysis is a novel approach which has been used to identify unique amino acids within a sub-family of phospholipases A2 (PLA2) in which the highly conserved active site residue Asp49 is substituted by Lys (Lys49-PLA2s). Although Lys49-PLA2s do not bind the catalytic co-factor Ca2+ and possess extremely low catalytic activity, they demonstrate a Ca2+-independent membrane damaging activity through a poorly understood mechanism, which does not involve lipid hydrolysis. Additionally, Lys49-PLA2s possess combined myotoxic, oedema forming and cardiotoxic pharmacological activities, however the structural basis of these varied functions is largely unknown. Using the 'SequenceSpace' analysis we have identified nine residues highly unique to the Lys49-PLA2 sub-family, which are grouped in three amino acid clusters in the active site, hydrophobic substrate binding channel and homodimer interface regions. These three highly specific residue clusters may have relevance for the Ca2+-independent membrane damaging activity. Of a further 15 less stringently conserved residues, nine are located in two additional clusters which are well isolated from the active site region. The less strictly conserved clusters have been used in predictive sequence searches to correlate amino acid patterns in other venom PLA2s with their pharmacological activities, and motifs for presynaptic and combined toxicities are proposed.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Argentina at the 2002 Winter Olympics
Argentina competed at the 2002 Winter Olympics in Salt Lake City, United States.
Alpine skiing
Men
Men's combined
Women
Women's combined
Biathlon
Men
Women
1 A penalty loop of 150 metres had to be skied per missed target.
3 One minute added per missed target.
Freestyle skiing
Men
Luge
Men
Skeleton
Men
References
Official Olympic Reports
Olympic Winter Games 2002, full results by sports-reference.com
Category:Nations at the 2002 Winter Olympics
2002 Winter Olympics
O
| null |
minipile
|
NaturalLanguage
|
mit
| null |
[Imaging of genetic aspects of Parkinson's disease].
Although the mechanisms which cause Parkinson's disease (PD) are still poorly understood, research on monogenic forms of PD have demonstrated a significant genetic contribution to its etiology. Monogenic forms of PD only account for a minority of cases but offer a unique avenue of research into the pathogenesis of PD. In this article the potential of structural and functional neuroimaging in monogenic forms to provide general insights into the pathophysiology of PD, including the more common idiopathic disease is reviewed. The review has a particular focus on neuroimaging of non-manifesting mutation carriers to study functional and structural changes in the brain at the asymptomatic stage of PD. This line of research has started to provide valuable insights into how the brain can cope with a latent nigrostriatal dopaminergic deficit and thereby delay the clinical onset of PD.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Your Body is a Magical Entity
Your body is a magical entity. It has all the knowledge you need now and forever to create a perfect life for you. Your body knows how not to be ill. It knows how to be in perfect wellness at all times.
Let me say that the individual cells of your body know that. When those cells come together to form YOU, the message sometimes gets lost in the translation. As you have embodied many, many times in your history, you have drawn to you the little imperfections that are found in your lifeform. One by one those imperfections over the years and centuries of your incarnations have coagulated to create a less than perfect body - the blueprint of which is carried forward from lifetime to lifetime. We know this is a bit much for most of you to grasp. However, we ask that you continue to read for we have much more to tell you.
These same imperfections which you believe you see in your physical bodies are also present in your mental and emotional bodies. It is as if each imperfection exists equally along the spectrum of density; i.e., first in the mental, then in the emotional and finally in the physical. If that is the case, how can these imperfections be restored to their intrinsic glory? How can your body, emotions and intellect all return to the perfection which is perfectly permeable to source energy?
Well, dear ones, that is why you are here on this planet. Your job is to figure this out. Actually, your first job is to recognize that your bodies are in this condition. Then your next job is to remember what your perfection feels like. Then your next step is to realize that you have it within your power to return what you consider to be an imperfect being to that state of perfection. And, finally, you must actually accomplish it.
What a lot there is for you to do!
If you will read back through the messages you have had from your greatest teachers throughout the centuries, you will see that they are all the same. Oh, there are some minor differences in translation from one language to another. But, in essence, the message has always been the same and will always continue to be the same. You, human, you have the ability to return your body, emotions and intellect to the perfect condition of source energy.
But "how" you continue to ask. How do I do that?
Have you noticed how many more methods of healing are appearing on your planet today than ever existed before? Have you noticed how much more subtle even your most invasive medical procedures have become in the past few years? Well, these things are happening for a reason. They are happening because they are the means, at least on the physical level, for you to return to your alignment with source energy.
As you heal your body, your emotions and your thoughts will heal themselves also. And vice versa. Heal your thoughts and your emotions and bodies will heal themselves. It is all connected. Every piece of you is connected to every other piece of you.
And as you heal, you will notice that the methods you choose to heal yourself will grow ever more subtle. When your imperfections are in the densest parts of your being, the methods of healing yourself must address those areas. As you become more and more like unto source energy, it takes much less effort on your part to move another step closer to that source.
So we encourage you to listen to yourself. If you have been doing the same things for years because those things make you feel good - or at least they did when you started doing them - notice if your body wants to do something more gentle and refined. Notice if your emotions carry less and less power to trouble you and notice especially if your thoughts are becoming more and more gentle on your emotions. All these things are signs of your approach to a reunion with source energy. Allow them to happen.
For example, if you used to want to run marathons and now you would prefer to do qi gong or yoga or tai chi, then do those things. Your body is now needing gentler forms of attention since it is in a more highly refined condition of being.
It is the same with your emotions. Perhaps wild parties used to please you but now you would prefer a simple walk in your neighborhood. Then take that action instead of going to the party.
Your bodies are telling your something. If your thoughts used to be those of anger and hatred but now you find yourself more in a state of compassion than not, accept that. You are moving closer and closer to source. Allow source to infuse you with its own energy. Allow the energy of source to light you up. Others will begin to notice and it will help them greatly to see this in you. We hope this has been sufficient for you. We encourage the development of your spiritual bodies and we applaud all of you who see it happening for yourself and recognize it for what it is. Many blessings.
This teaching was brought forward by my special group of spirit teachers known as Ananda.
Use of the quantum field, a.k.a. the primordial soup, a.k.a. the universal matrix, a.k.a the cosmic lattice, has become the new normal for emotional clearing. To learn more about this, CLICK HERE to download a FREE PDF of The Emotion Code. Reading this definitive book on the subject of stored emotional energy will really help you understand how easy it is to use the quantum field for emotional clearing work.
Site Disclaimer - Nothing on this site is meant to replace the advice of your doctor.
By using this site, you agree that any work we do together is solely for your own personal growth. It is not a therapeutic activity such as medicine, psychotherapy or counseling and is not intended as such.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Harper's to Shafer: You're All Wet on the Gitmo Story
And Shafer responds.
Harper's Senior Editor Luke Mitchell—who edited the Guantánamo feature by Scott Horton I criticized last week—takes sharp issue with that assessment in an e-mail that he sent me and has posted at the Harper's site. My response follows his e-mail.
From: Luke MitchellTo: Jack ShaferDate: Jan. 29, 2010
Dear Jack,
I just read your piece about Scott Horton's article, which I edited. I was surprised by the tone. I've enjoyed reading your skeptical take on various news reports in the past, but here I can't quite figure out what's got you so worked up. You have not refuted any of the facts in the piece, nor have you provided a specific example of what you call a "flight of logic." Basically, you seem to be arguing that the whole thing just sounds crazy.
Your first specific complaint regards the existence of Camp No. The place obviously exists. You reprinted a satellite photo of it. We have several named witnesses who describe it in detail. Since we published the story, we have also heard from another witness, a marine biologist who worked at Guantánamo, who corroborates those details. So what is it for? We don't know. As Scott reported, the people who saw it all thought it was a prison. Four of them were prison guards, so they have some expertise in the matter. But, as Scott also reported, they also don't know for sure.
You object that "Hickman further claims to have once heard a 'series of screams' emanating from Camp No, but Horton doesn't explain how Hickman got close enough to the restricted space to hear the screams." It is difficult to parse your concern here. Are you suggesting that Hickman was lying about what he heard? That Horton was lying about what Hickman told him? That Hickman was telling the truth, but Horton somehow subverted that truth by failing to explain precisely how Hickman went about approaching the camp? (Or the "restricted space," as you have oddly deemed it.)
So far, this doesn't amount to much more than you wiggling your eyebrows and saying "sounds fishy to me." Next, you expand your vague concern about Camp No's accessibility into an accusation that Scott has designs to "lull" his readers, who are "gullible," into thinking that Camp No not only exists, but is used as a torture site. As I mentioned, Scott does report many facts that support the proposition that Camp No exists. He also reports many facts that—and here I use the word Scott uses throughout the piece—suggest torture took place there. I think most readers understand the difference between "suggest" and "prove." I assume you do as well, and so it's not clear to me what sort of Mephistophelian game you think is afoot.
Your next complaint has to do with the illogical nature of the proceedings of the night of June 9. First, you envision a nutty scenario—"if you were going to torture prisoners to the point of death in interrogations, would you really draw three prisoners from the same cell block, inside the same hour, for that punishment"—and then you knock it down by noting how nutty it is. Perhaps it is nutty. But that scenario is your own invention.
It is worth noting that Human Rights First and others have reported many cases in which prisoners have died in connection with the use of Justice Department-sanctioned "harsh interrogation techniques." Last summer, General Barry McCaffrey expressed this quite forthrightly. "We tortured people unmercifully," he said. "We probably murdered dozens of them during the course of that, both the armed forces and the C.I.A." So the phenomenon is hardly unknown.
Nonetheless, and contrary to your implication, Scott does not advance a clear theory as to how the deaths occurred. He questions the validity of the official government narrative, and he points to significant new evidence that suggests that the deaths occurred at another location and in other circumstances. But too few facts are available to say for certain what happened that night, which is precisely why Scott does not attempt to do so.
Your next complaint brings your larger concern into focus. You suggest that, even if "the CIA is capable of such a crime," readers should not be asked to believe that "the entire U.S. government—across two administrations—is willing to devote its energies to a cover-up." And with that suggestion—again drawn entirely from your own imagination—we have reached the dread conspiracy moment. I say dread because calling a news report a conspiracy theory is the press critic's version of comparing a politician to Hitler. It's less than a straw man. It's not even an argument, and so it is nearly impossible to refute.
Still, let's be clear: Scott has not posited a grand conspiracy. What he has reported is that there is strong evidence to suggest that some—perhaps even many—government officials are not telling the truth about what happened on June 9, 2006. Now perhaps your life experience has taught you that it's just la-la crazy to think government officials would ever lie about anything. But I hope you can at least recognize that there is some precedent for an alternative view.
Finally, taking your leave with a somewhat unpersuasive "I could go on," you recite many other elements of the piece, but rather than dispute them, you simply bracket them with dismissive terms like "Horton seems to believe," and "as Horton puts it," and—inevitably—"allegedly." This is just a mockery of serious criticism.
It's fun to knock down stories, and when the stories are false, it's important to knock them down. But Scott's story is not false. You may think he has put too much spin on the presentation. Fine. But your attempt to transform a stylistic concern into a factual concern is a disservice to your readers.
As Scott wrote, "Nearly 200 men remain imprisoned at Guantánamo. In June 2009, six months after Barack Obama took office, one of them, a thirty-one-year-old Yemeni named Muhammed Abdallah Salih, was found dead in his cell. The exact circumstances of his death, like those of the deaths of the three men from Alpha Block, remain uncertain." These facts alone create a certain urgency about this case that is belied by your normally welcome insouciance.
Scott does not claim to know precisely what happened at Guantánamo in 2006. But we believe he has made a strong case that the NCIS narrative is contradictory on its own terms, that new witnesses have presented compelling new testimony, and that this new testimony cries out for further investigation.
We stand by his story.
You obviously are welcome to quote from this, or to write or call with any follow-up questions.
All best,
Luke MitchellSenior EditorHarper's Magazine
Jack Shafer replies:
Luke,
Advertisement
Thanks for the kind words at the top of your letter, but I think you're wrong about my "tone." If you've read much of my work, you would know that Horton's piece didn't get me "worked up" at all. I did my best to describe Horton's shortcomings in a measured fashion. Now, on to the main event.
You're deliberately misreading my column if you think I dispute the existence of a Guantánamo building that the Horton piece repeatedly calls "Camp No." I can see it in the map you published, and I can see it on Google Earth. What I dispute is 1) that the building's name is "Camp No" and 2) that we have good reason to believe that it is a prison or torture chamber.
The article provides no evidence, outside of the speculation of four prison guards, that the mysterious building houses a prison. As you point out in your note, the guards "don't know for sure" what the building is used for. If they don't know, how can Horton build such an elaborate narrative about prisoners being transported to the facility alive and returned to Camp 1 dead on the evening of June 9, 2006? I consider this a flight of logic, and you should, too.
A just-published Horton dispatch, to which you refer, quotes a marine biologist who says he saw the mysterious structure in February 2004 while working on a project for the Pentagon. "[I]t looked a lot like the other prison camps I had seen," the biologist tells Horton, adding, "There didn't seem to be any windows in the facility."
Advertisement
If the building is a prison without windows, as the biologist surmises, how do we reconcile his description account with the account provided by Hickman, who tells Horton that "he heard a 'series of screams' from within the compound" one time that he stopped by. If the building has no windows, it's hard to imagine Hickman hearing a series of screams from inside. If guards kept doors ajar while they tortured the prisoners—or tortured prisoners outside—I suppose you could hear their screams. I'd love to know if Hickman, who says he was in earshot of the building, could tell us whether it has windows.
And seeing as you asked, I'm not calling anybody a liar. Do you think that the dozens of soldiers, sailors, and civilians who gave sworn testimony to NCIS investigators (large PDF, heavily redacted), testimony that contradicts Horton's findings, have perjured themselves? Are they liars?
Your note falls into a sinkhole of sophistry when you claim that Horton has no "clear theory" of how the prisoners died, and that he's only "suggesting" that the prisoners were tortured at the building he calls "Camp No." Are we reading the same piece? The Harper's headline places the word suicide in quotation marks ("The Guantánamo 'Suicides': A Camp Delta sergeant blows the whistle") to advance his idea that the prisoners did not take their own lives. Horton repeatedly refers to evidence or the possibility of a "cover-up," and he endorses the Seton Hall study (PDF) to claim that the "official story of the prisoners' deaths was full of unacknowledged contradictions, and the centerpiece of the report—a reconstruction of the events—was simply unbelievable."
What's this stuff about "conspiracy theory"? Others have called Horton's work conspiracy theory, but nowhere do I describe his work that way. Also, in using the words and phrases "seems to believe," "puts it," and "allegedly," I intended no mockery.
Advertisement
Since my piece appeared, several readers have called my attention to what they regard as a glaring error in Horton's feature, one that undermines his counter-narrative of the events of June 9, 2006. One of those readers was Dwight Sullivan, who served as the chief defense counsel in the Office of Military Commissions from August 2005 through August 2007. In that capacity, Sullivan drove from the naval station portion of Guantánamo Bay to the detention camps many times. Via e-mail he writes:
Mr. Horton's article states that "past the perimeter checkpoint known as ACP Roosevelt, there were only two destinations. One was a beach where soldiers went to swim. The other was Camp No." That statement isn't true. The checkpoint was on the main road between the Naval Station and the detention camps. After driving from the detention camps and passing the checkpoint, every facility on the Naval Station was a potential destination. This includes facilities to which detainees were sometimes transported, including the main hospital and the military commission building.
The upshot of Horton's geographical error? The "paddy wagon" that Hickman says he saw depart from the camp with prisoners could have gone to destinations other than "Camp No." Even the paddy wagon's third trip, that Hickman has turning left toward "Camp No" after clearing the checkpoint, could have continued past the turn-off for "Camp No," if I understand the map correctly. (For an additional point of reference, see this link to the map Harper's annotated.)
Did you serve at Guantánamo. Let me hear from you, too. Send e-mail to [email protected] or take orders from my Twitter feed. (E-mail may be quoted by name in "The Fray," Slate's readers' forum; in a future article; or elsewhere unless the writer stipulates otherwise. Permanent disclosure: Slate is owned by the Washington Post Co.)
Track my errors: This hand-built RSS feed will ring every time Slate runs a "Press Box" correction. For e-mail notification of errors in this specific column, type the word Horton in the subject head of an e-mail message and send it to [email protected].
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Rangers’ players boycott trip to Zambia over unpaid allowance
Players of Ranger International refused today to board an Ethiopian airline flight to Zambia, where they are scheduled to play the second leg of the Confederation Cup.
The players were protesting the non-payment of their travelling allowances.
A News Agency of Nigeria (NAN) correspondent who was at the Airport reports that trouble started when the players were told by management of the team that they will receive their allowance when they return to Nigeria.
The information made the players to revolt and insist that they were not going to honour the match, unless they were paid.
Some of the players said that the management still owed them the match allowance due them when they met JS Suoara of Algeria in February.
News Agency of Nigeria reports that the match against JS Souara was played in February 2017 and the first leg ended 1-1 draw in Algeria with 0-0 draw in Enugu.
All efforts by the General Manager/CEO of the club, Christian Chukwu and the Team manager, Amaobi Ezeaku, to convince the players to embark on the journey fell on deaf ears.
The management was convinced that the players meant business when they jumped into their respective vehicles and left the Airport.
However, on receiving information about the protest, the state Commissioner for Youth and Sports, Charles Ndukwe drew the attention of the governor, Ifeanyi Ugwuanyi, who immediately released fund for the settlement of the allowances.
But his intervention came late. The players could not make the flight as the plane had departed.
The development forced the management to shift the journey till tomorrow Friday, when the 28 -man contingent would travel to Lagos en-route Zambia for the encounter.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
20141128 - Week in Photos November 23 to 29, 2014
Monica Synett - [email protected]
Rosary junior Katie Rentz catches a breath in the 100 yard butterfly during the IHSA Girls Swimming State Championships hosted by Evanston Township High School on Saturday, November 22, 2014. Rentz finished 11th in the race with 57.00.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Best for women and children, one size fits all. Pre- or post-dive, these bad boys’ll keep your mitts toasty warm. Smoko Toast Handwarmers are squishy, cute and comfy. Just plug them into the nearest USB port, and switch them on. In just a few moments, your hands are wonderfully toasty. You can even type on a keyboard with them on if you’re into downloading your dive info from dive computer to laptop right onboard in the cold elements. Size: 5.5x5.5inches/Cord length 57inches, Two Adjustable Wrist Straps. Smoko thought of it all!
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Identification of prognostic factors associated with early mortality after surgical resection for pancreatic cancer--under-analysis of cumulative survival curve.
The cumulative survival curve after surgery for advanced pancreatic cancer is characterized by a steep downward slope in the early postoperative period. The aim of this investigation was to identify the characteristics associated with early mortality in patients undergoing pancreatic resection for pancreatic cancer. Thirty-seven patients with extended radical pancreatectomy combined with intraoperative radiation therapy were studied. The cumulative survival curve in this series was depicted using the Kaplan-Meier method. Assuming that there were two distinct survival curves, below and above the breakpoint, each part of the curve was modeled as an exponential distribution. Three parameters, the breakpoint, the high hazard rate below the breakpoint, and the low hazard rate above the breakpoint were estimated by the maximum likelihood method. Prognostic factors associated with early mortality after surgery were evaluated using univariate and multivariate Cox proportional hazards regression analyses. The breakpoint of the survival curve was estimated at 41 months. The short-survival group (SSG) was defined as deceased earlier than 41 months after surgery, and included 31 patients (83.8 %). The long-survival patient group (LSG) consisted of 6 patients who were alive more than 41 months after surgery. Eighteen SSG patients (58.1 %) died of hepatic metastases, whereas no LSG patients died of hepatic metastases. Abdominal pain and/or back pain during clinical course was identified by multivariate analysis as a prognostic factor for patients undergoing pancreatic resection. The high hazard rate in the early postoperative period was closely linked with death due to liver metastases. The preoperative presence of local pain was a prognostic factor associated with early mortality.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
408 F.2d 684
Edith Portis BAUGHN, Administratrix of the Estate of WilliamEdwin Baughn, Deceased, Appellant,v.SEABOARD AIR LINE RAILROAD COMPANY, Appellee.
No. 13088.
United States Court of Appeals Fourth Circuit.
Argued April 9, 1969.Decided April 18, 1969.
Clyde C. Randolph, Jr., Winston-Salem, N.C., for appellant.
Fred Bynum, Jr., Rockingham, N.C. (James D. Blount, Jr., and Leath, Bynum, Blount & Hinson, Rockingham, N.C., on brief), for appellee.
Before HAYNSWORTH, Chief Judge, and BRYAN and WINTER, Circuit Judges.
PER CURIAM:
1
To recover damages for the death of William Edwin Baughn when struck by a train of the Seaboard Air Lone Railroad Company on June 1, 1966, while driving a truck over a highway crossing of the tracks a few miles east of Lilesville, North Carolina, his adminstratrix sued the railroad. The District Court directed a verdict for the defendant and dismissed the action, for the decedent's contributory negligence.
2
In his opinion, Baughn v. Seaboard Air Line Railroad Company, 291 F.Supp. 425 (M.D.N.C.1968), the District Judge recounts closely and with clarity the proofs upon the tragic event. The issue of contributory negligence is decided in this context and in the frame of the law of North Carolina. Since our review discloses no substantial error in the premises or the final determination, we cannot overrule the dismissal.
3
Affirmed.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
LISTEN TO ARTICLE 1:54 SHARE THIS ARTICLE Share Tweet Post Email
Source: Getty Images Source: Getty Images
Want the lowdown on European markets? In your inbox before the open, every day. Sign up here.
As opposition leader Jeremy Corbyn tries to lure voters with a plan to return much of Britain’s broadband network to state control, events on the other side of the world provide a cautionary tale.
Australia offers the most ambitious example of a government-led internet rollout in a developed economy and the project has become a byword for mismanagement and delays.
Strategy reversals and cost overruns have left the country languishing at 61st in a global ranking of internet speeds, behind Belize, Paraguay and some former Soviet bloc nations.
U.K. Labour Plans to Nationalize BT’s Broadband Unit
The Australian government merged private fixed-line assets into state broadband monopoly NBN Co. on a promise to connect 93% of homes with high-speed optic fiber lines. When that proved too expensive, it opted for a mix of technologies. The legacy is a frustrated public who often find their mobile phones now offer faster connections than landlines.
There’s a reason why only one other country has come close to doing what the U.K.’s Labour Party is planning, said Matthew Howett, founder and principal analyst at policy research firm Assembly. “It’s hard, expensive and fraught with difficulty,” he said in an email.
A British government-commissioned report last year found that Australia’s plans are set to take three years longer than planned and require an extra A$73 billion ($45 billion).
Singapore took a state-led approach and managed to roll out 100% full-fiber coverage in seven years, although that was made easier by its much smaller and denser population. New Zealand, which sits in 23rd place for broadband speeds, used private regional franchises backed up by state subsidies.
Australia now wants to sell NBN back to the private sector. Even that may prove problematic.
“There’s growing evidence that 5G mobile connections will bypass the NBN network and that’s now threatening the privatization plan,” said Ian Martin, a senior telecom analyst at New Street Research.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Q:
Change the width of the Dialog window of Android app dynamically
I am creating an Android application. In that application, I am using
a Dialog window to display HTML contents which come from the sever
side. I created the Dialog window using below method.
private void displayDialogWindow(String html) {
// here I have created new dialog window
dialog = new Dialog(cordova.getActivity(),
android.R.style.Theme_NoTitleBar);
dialog.getWindow().getAttributes().windowAnimations =
android.R.style.Animation_Dialog;
dialog.requestWindowFeature(Window.FEATURE_NO_TITLE);
dialog.setCancelable(true);
// creates a new webview to display HTML and attached to
dialog to display
webview = new WebView(cordova.getContext());
webview.setLayoutParams(new LinearLayout.LayoutParams(
LayoutParams.FILL_PARENT, LayoutParams.FILL_PARENT));
webview.setWebChromeClient(new WebChromeClient());
webview.requestFocusFromTouch();
final String mimeType2 = "text/html";
final String encoding2 = "UTF-8";
webview.loadDataWithBaseURL("", html, mimeType2, encoding2, "");
dialog.setContentView(webview);
// get the screen size of the device
WindowManager wm = (WindowManager) ctx
.getSystemService(Context.WINDOW_SERVICE);
Display display = wm.getDefaultDisplay();
Point size = new Point();
display.getSize(size);
int width = size.x;
int height = size.y;
// set appropriate sizes to the dialog window whether there is
side menu or not
lp.copyFrom(dialog2.getWindow().getAttributes());
lp.width = width ;
lp.height = height - 100;
dialog.getWindow().setAttributes(lp2);
dialog.show();
}
Using above method I could successfully create a Dialog window.
And I am creating the Dialog window using a Thread,
Runnable runnable = new Runnable() {
public void run() {
displayDialogWindow(html);
}
};
But after display the Dialog window once, I want to change the width
of the Dialog window dynamically. (eg. when I click on another button
of my app, the width of the window should be decreased and come to the
original width when another button is pressed).
Can any one help me to change the width of the Dialog window dynamically?
A:
I think that you have to "recreat" your dialog everytime you want to change it.
After yourDialog.show() method, you can insert this line:
yourDialog.getWindow().setLayout((6 * width)/7, LayoutParams.WRAP_CONTENT);
This allow you to resize your Dialog window and it's content.
Of course width and height are the size of your screen.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
The turtle appeared to be 8 or 9 years old, and weighed 365 grams.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
There are so many changes in child safety seats nationwide that it is often difficult to know what is the “right” thing to do for your child. Here’s a great site that shows you EXACTLY where your child should be, carseat-wise.
Visit boosterseat.gov to read the latest regulations on car safety seats. You can even use their “quick check” on the left margin to enter your child’s age, height and weight and find the ideal seat for your child!
As it turns out, my kids are actually within the recommendations. My son (just turned is about to relinquish the booster as he is 52 inches tall and weighs over 70 pounds. And my younger son – 2.5 years – is right at the cusp of changing into a booster seat.
Something this website does not point out is a little number called the EXPIRATION DATE. Car seats are only rated for 5 years of use; standards change, materials change, and to keep your child the safest, it is best to heed these dates. Pay attention to these dates (my younger son’s seat has recently expired, so thank goodness I was going to have to buy him a new one anyway for his next step up!).
Children should ALWAYS be buckled into a car safety seat unless they are old enough and big enough to do without. And, once they reach the point that they can leave the booster seat behind and seat in the “regular” seat, make sure they are still buckling up, as you should be, too. In our house, the car does not leave the driveway unless all the seatbelts are buckled. Please pay attention to the recommended seats for your children. That’s our world’s future you’re buckling back there.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
A bead based multiplex immunoassay detects Piscine orthoreovirus specific antibodies in Atlantic salmon (Salmo salar).
Future growth in aquaculture relies strongly on the control of diseases and pathogens. Vaccination has been a successful strategy for obtaining control of bacterial diseases in fish, but for viral diseases, vaccine development has been more challenging. Effective long-term protection against viral infections is not yet fully understood for fish, and in addition, optimal tools to monitor adaptive immunity are limited. Assays that can detect specific antibodies produced in response to viral infection in fish are still in their early development. Multiplex bead based assays have many advantages over traditional assays, since they are more sensitive and allow detection of multiple antigen-specific antibodies simultaneously in very small amounts of plasma or serum. In the present study, a bead based assay have been developed for detection of plasma IgM directed against Piscine orthoreovirus (PRV), the virus associated with the disease Heart and skeletal muscle inflammation (HSMI) in farmed Atlantic salmon. Using recombinant PRV proteins coated on beads, antibodies targeting the structural outer capsid protein μ1 and the non-structural protein μNS were detected. Results from a PRV cohabitation challenge trial indicated that the antibody production was initiated approximately two weeks after the peak phase of PRV infection, coinciding with typical HSMI pathology. Thereafter, the antibody production increased while the epicardial inflammation became less prominent. In conclusion, the novel assay can detect PRV-specific antibodies that may play a role in viral defence. The bead-based immunoassay represents a valuable tool for studies on HSMI and possibly other diseases in aquaculture.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
GETTY IMAGES Ross Brawn said there is still hope for Michael Schumacher
FREE now and never miss the top politics stories again. SUBSCRIBE Invalid email Sign up fornow and never miss the top politics stories again. We will use your email address only for sending you newsletters. Please see our Privacy Notice for details of your data protection rights.
Ross Brawn, who ran the Ferrari F1 team during the German legend's time in the famous red car, says he keeps in touch with Schumacher's family. He has visited Formula One's most successful driver a number of times and while Mr Brawn said the seven times world champion's recovery progress is "slow" he remains positives. Schumacher is currently involved in a battle to recover from catastrophic brain injuries suffered during a ski accident in 2013.
GETTY IMAGES Michael Schumacher won five world championships in his Ferrari
We just keep praying every day that he’ll recover to a stage where…it's slow, but there's always hope. Ross Brawn
Mr Brawn, the former team principal of Ferrari and Mercedes, said: "I do keep in touch [with Schumacher’s family], but we try and keep a balance of going to see him against calling, and not being a pain. I’ve been to see him a few times; Corinna, his wife, calls me occasionally and keeps me updated. "We just keep praying every day that he’ll recover to a stage where…it's slow, but there's always hope."
GETTY IMAGES Michael Schumacher was Formula One World Champion seven times
Mr Brawn has been instrumental throughout Schumacher's glittering Formula One career. He served as technical director for the Benetton team from 1991 until 1996, where Schumacher won his first two world championships. He then joined Ferrari in 1997 - along with Schumacher - and helped him to win five titles with the Scudderia. Schumacher's career saw him amass a record seven world titles and the most amount of wins (91) ever for a driver.
GETTY IMAGES Michael Schumacher suffered devastating injuries in a skiing accident
The F1 legend has been battling for his life ever since he suffered catastrophic head injuries in December 2013 during a ski accident. While going across an area that was not on piste, Schumacher hit his head on a rock and was needed emergency brain surgery. He was placed in a medically induced coma for nearly six months but then in June 2014 was discharged from hospital and has since been receiving medical care at home.
Michael Schumacher F1 Legend Thu, July 30, 2009 Play slideshow 1 of 12
However, his recovery and current condition has been shrouded in secrecy - with official updates few and far between. Schumacher has reportedly been left paralysed following the tragic ski accident but those claims are unconfirmed. Earlier this month Express.co.uk reported that Schumacher was running up a care bill of £115,000 a week, with the total costs for the F1 racing legend hitting £12million. He remains under the care of a 15-strong team of physicians, nurses and medical technologists at his custom-built mansion home on the shores of Lake Geneva in Switzerland.
GETTY IMAGES Ross Brawn says there's still hope for Michael Schumacher
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Motor vehicles such as electric vehicles or hybrid vehicles have been widely used in recent years. Motor vehicles require that a battery with a large capacity be mounted thereon, because a battery with a larger capacity provides a longer cruising range. It is therefore proposed to mount a battery pack that stores a plurality of battery modules under the floor of an automobile. JP 2011-121483 A, for example, proposes a structure in which a battery pack including battery frames extending in the vehicle forward-rearward direction is mounted under a floor panel such that portions of the battery frames on the vehicle rear end side are fastened, via a floor cross member, to rear side members.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
J-S49044-18
2018 PA Super 288
COMMONWEALTH OF PENNSYLVANIA : IN THE SUPERIOR COURT OF
: PENNSYLVANIA
:
v. :
:
:
ELIUD MONTANEZ-CASTRO :
:
Appellant : No. 462 MDA 2018
Appeal from the Judgment of Sentence December 28, 2017
In the Court of Common Pleas of Dauphin County Criminal Division at
No(s): CP-22-CR-0003013-2016
BEFORE: SHOGAN, J., STABILE, J., and STEVENS*, P.J.E.
OPINION BY STEVENS, P.J.E.: FILED OCTOBER 22, 2018
Appellant, Eliud Montanez-Castro, appeals from the judgment of
sentence entered in the Court of Common Pleas of Dauphin County, which,
sitting as finder of fact in his non-jury trial, found him guilty of luring a child
into a motor vehicle, disorderly conduct, and harassment.1 Sentenced to
serve an eleven-and-one-half to 23-month sentence of incarceration, to be
followed by one year of probation, Appellant challenges the sufficiency of the
evidence with respect to his conviction for luring a child into a motor vehicle.
We affirm.
The trial court sets forth the pertinent facts and procedural history, as
follows:
On April 15, 2016, the two minors (T.G. and M.R.) were walking
to school. Neither recalled inclement weather or any sort of
natural disaster that would prompt anyone to offer a ride. As they
____________________________________________
1 18 Pa.C.S. §§ 2910(a), 5503(a)(4), and 2709(3), respectively.
____________________________________
* Former Justice specially assigned to the Superior Court.
J-S49044-18
were walking along South Harrisburg Street, a white car sharply
turned left from Walnut Street onto South Harrisburg Street. It
pulled up to the curb beside them, with the driver’s side window
closest. The window was open and the driver, Appellant, asked
them to pick a number. The minors stopped walking and after
being confused, each picked a number. After T.G. picked a
number, Appellant said “Mommy, you were right” and that she
won a prize. T.G. thought this was all odd.
Appellant held out a soda can and water bottle with no label and
told her to pick on[e]. T.G. refused. Both girls recalled that
[Appellant] held the drinks out with bent elbows, not arms
extended, and was fairly close [just several feet away]. T.G. and
M.R. recognized that T.G. would have had to approach the car,
though, to take one of the drinks. T.G. and M.R. were nervous
and scared after this interaction and walked quickly to school.
Appellant never asked her to enter the car or go anywhere with
him. He never asked them to approach closer, he never
threatened them, he never commanded or directed them to do
anything, and he never opened the door. Appellant did not follow
them. Upon arriving at school, T.G. told the principal what had
happened.
Sometime later, T.G. viewed a photo array with Detective
Robbins. He showed her pictures one at a time and she
immediately identified picture six as the man who had approached
her. M.R. also viewed a photo array complied [sic] by Detective
Morris. He used the same technique as [Detective] Robbins and
showed M.R. pictures one by one until she identified one as the
man she had seen in the car.
M.R. recalled that some days later, she and T.G. were on the porch
when T.G. pointed out a man to her. M.R. looked and recognized
Appellant on the sidewalk. He looked up at them and they went
inside because they were scared.
Counsel presented a stipulation that Appellant did not have the
expressed or implied permission of any parent or guardian of the
victims in this case to give them a ride anywhere.
***
Following a trial by judge on [October 18, 2017], Appellant was
found guilty [of all charges and sentenced as indicated, supra].
-2-
J-S49044-18
On March 6, 2018, [the trial court] received a timely Notice of
Appeal filed with the Superior Court of Pennsylvania. [The trial
court] ordered Appellant on March 7, 2018, to file a concise
statement of matters complained of on appeal pursuant to
Pa.R.A.P. 1925(b). Appellant complied with said Order on March
26, 2018.
Trial Court Opinion, 8/6/18, at 1-3.
Appellant presents one question for our consideration:
I. [WAS] THE EVIDENCE PRESENTED AT TRIAL [ ]
INSUFFICIENT TO CONVICT APPELLANT OF THE
CRIME OF LURING A CHILD INTO A MOTOR VEHICLE
AND DISORDERLY CONDUCT WHEN THE APPELLANT
MERELY OFFERED A SODA TO A GIRL(S) [SIC]
THROUGH A PASSENGER WINDOW OF HIS VEHICLE
BUT NEVER OFFERED A RIDE TO THE GIRL(S) OR
PULLED THEM CLOSER TO HIS VEHICLE[?]
Appellant’s brief, at 4.
Our standard of review for challenges to the sufficiency of the evidence
is well-settled:
The standard we apply in reviewing the sufficiency of the evidence
is whether viewing all the evidence admitted at trial in the light
most favorable to the verdict winner, there is sufficient evidence
to enable the fact-finder to find every element of the crime beyond
a reasonable doubt. In applying [the above] test, we may not
weigh the evidence and substitute our judgment for the fact-
finder. In addition, we note that the facts and circumstances
established by the Commonwealth need not preclude every
possibility of innocence. Any doubts regarding a defendant's guilt
may be resolved by the fact-finder unless the evidence is so weak
and inconclusive that as a matter of law no probability of fact may
be drawn from the combined circumstances. The Commonwealth
may sustain its burden of proving every element of the crime
beyond a reasonable doubt by means of wholly circumstantial
evidence. Moreover, in applying the above test, the entire record
must be evaluated and all evidence actually received must be
considered. Finally, the [finder] of fact while passing upon the
-3-
J-S49044-18
credibility of witnesses and the weight of the evidence produced,
is free to believe all, part or none of the evidence.
Commonwealth v. Harden, 103 A.3d 107, 111 (Pa.Super. 2014) (citation
omitted).
Appellant argues that evidence failed to support his conviction on the
charge of luring a child into a motor vehicle. Merely talking with the girls and
offering them drinks from his car window, without ever inviting or
commanding them to enter his vehicle, fails to satisfy the first evidentiary
requirement of attempting to lure a child into a vehicle, he submits. On this
point, he elaborates:
The Commonwealth failed to establish that Appellant lured or
attempted to lure either lady into his motor vehicle. In this case,
neither lady called 911. Neither lady testified that he said “get in
the car” or offered them to get into the vehicle [sic] or that he
pulled them into the car. He never opened the rear car door for
them to get in. He did not even say “come here.” He did not
beckon them. He did not threaten them. He did not command
them to come closer or open the door. At no time did he even
touch them.
Appellant’s brief, at 9.2
The pertinent statute provides: “(a) Offense.—Unless the
circumstances reasonably indicate that the child is in need of assistance, a
person who lures or attempts to lure a child into a motor vehicle or structure
____________________________________________
2 Throughout the “Brief for Appellant,” counsel repeatedly refers to the two
minor girls in question as “ladies.” Assigning an adult title to minors subjected
to an offense that frequently involves the actual or attempted sexual abuse of
a child is unacceptable, particularly where clear sexual overtones attended
Appellant’s conduct toward the girls. Therefore, we strongly admonish counsel
for what was, at best, a careless description of the minor victims in this case.
-4-
J-S49044-18
without the consent, express or implied, of the child's parent or guardian
commits an offense.” 18 Pa.C.S. § 2910(a). As stated by our Supreme Court:
Section 2910 ... sets forth three requirements the Commonwealth
must establish beyond a reasonable doubt to convict an individual
of the offense of attempted luring of a child into a motor vehicle:
(1) the individual attempted to lure a child into a motor vehicle;
(2) without the express or implied consent of the child's parent or
guardian; and (3) under circumstances which did not reasonably
indicate the child is in need of assistance.
Commonwealth v. Hart, 28 A.3d 898, 908–09 (Pa. 2011). The Court has
explained further that “a ‘lure’ involves the making of a promise of pleasure
or gain, the furnishing of a temptation or enticement, or the performance of
some other affirmative act calculated to strongly induce another individual to
take a particular action, usually and most often likely to result in his or her
harm.” Id at 909.
Here, two ninth-grade girls walking to school witnessed Appellant
abruptly turn his vehicle off the course he was traveling and drive onto the
curb alongside where they were walking. He immediately stopped the girls
and engaged them in conversation, invited each to “pick a number,” and
offered to the girl whom he declared the winner a bottled drink as her prize.
In order to claim her prize, however, the girl, whom Appellant was now
calling “Mommy,” was required to walk up to Appellant’s car window. There
Appellant awaited, holding the bottle with his arm bent inward toward himself
so that the girl would have to come right to his side to retrieve it. The girls
were scared and nervous at this point, and they walked away.
-5-
J-S49044-18
When viewed in a light most favorable to the Commonwealth as verdict
winner, the totality of circumstances allows for the reasonable inference that
Appellant did not stop two random girls on their way to school to give them a
drink and send them on their way. Appellant was, instead, attempting to
entice at least one girl to come within his area of control as he waited in his
car, and he used his apparent brand of charm and a bottled drink—kept closely
to his side—as a lure to achieve this end.
Indeed, Appellant neither knew the girls nor had any reason to believe
they were in need of assistance. Yet, he stopped them anyway. He
immediately focused on one girl, declared her a winner, gave her the
suggestive pet name “Mommy,” and invited her to come to him where he sat
in his car and claim her prize.
Therefore, it was reasonable for the finder of fact to conclude, beyond a
reasonable doubt, that Appellant’s affirmative actions, manipulative and
suggestive words, and enticements were designed to gain the minor girl’s
entry into his car to her own detriment. Accordingly, evidence adduced at
trial sufficed to establish that Appellant committed the offense of attempting
to lure a child into his vehicle.3
Judgment of sentence affirmed.
____________________________________________
3 Although Appellant’s “Statement of the Question Presented” also raises a
sufficiency challenge to his disorderly conduct conviction, he presents no
argument to support this claim. Therefore, he has waived this claim. See
Pa.R.A.P. 2119(c); Commonwealth v. Plante, 914 A.2d 916 (Pa.Super.
2006) (failure to develop an argument with citation to authority waives the
issue on review).
-6-
J-S49044-18
Judgment Entered.
Joseph D. Seletyn, Esq.
Prothonotary
Date: 10/22/2018
-7-
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Q:
Handle hidden popup in remote powershell msiexec installation
I need to perform upgrading of applications remotely via powershell. Uninstall works fine, but because for installation, there is a popup that requires me to click on "ok" before proceeding. While I could shut that down in the local installation instance, I can't do that for remote installations as that popup is hidden (not shown). My installation thus fails. Here's my script:
Enter-PSSession "computername" -Auth CredSSP -cred username
(Start-Process -FilePath "msiexec.exe" -ArgumentList "/i D:\path\installer.msi /l*vx D:\path\installation.log /q" -Wait -Passthru).ExitCode
The execution of the above script hangs because of the hidden popup. Any ideas on how I can suppress the hidden popup or send a 'enter' instruction, preferably without using psexec?
Resources I already referenced: 1 | 2 | 3 | 4 | 5 | 6 |
A:
I solved my problem. Maybe it was a corner case that the solution might work only for my case, but I'll share it in case others wish to try out (/LM/W3SVC/2 and <appPool name> and other paths need to be changed accordingly to your setup):
Old Script
msiexec /i D:\path\installer.msi TARGETAPPPOOL=<appPool name> TARGETSITE=/LM/W3SVC/2 /l*vx D:\path\installation.log /quiet
The popup persists, even with /quiet or /qn. It's made worse when it's hidden if executed remotely. But I chanced upon setting the TARGETDIR and the popup was acknowledged automatically and taken care of:
New Script
msiexec /i D:\path\installer.msi TARGETAPPPOOL=<appPool name> TARGETDIR=D:\path TARGETSITE=/LM/W3SVC/2 /l*vx D:\path\installation.log /quiet
The new property attribute seems unrelated to my problem, but nonetheless solved it.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Giftedness is wonderful in many ways, but gifted children may also have characteristics that can affect their social and emotional development. To understand your gifted child, it's a good idea to see how your child's giftedness can influence his or her behavior.
Asynchronous Development Issues
Advanced intellectual skills are not always accompanied by advanced social and emotional skills. When skills develop at different speeds, development is described as "asynchronous." In some cases, gifted children can run into problems when their intellectual power is not matched by their other abilities. For example:
Gifted children can intellectually understand abstract concepts but may be unable to deal with those concepts emotionally, leading to intense concerns about death, the future, sex, and other advanced issues.
Gifted children's physical development may lead to an inability to complete a task they are capable of intellectually envisioning. This can lead to extreme frustration and acting out. (Perfectionism may play a role in this frustration as well.)
A gifted child may be able to participate in adult conversations about issues such as global warming or world hunger one minute and the next minute cry and whine because a sibling took a favorite toy. This may confuse adults, and cause an overreaction to age-appropriate behavior.
Advanced Verbal and Reasoning Ability
While gifted children are capable of reading, speaking, and even reasoning above grade level, those abilities may not always be used in positive ways.
For instance, gifted children can be argumentative and/or manipulative. (Adults often remark that these children are little lawyers.) Parents and other adults need to remember that, although credit should be given for logical and convincing arguments, a child is still a child and requires appropriate discipline, no matter how clever or cute the behavior may look.
Children who see that they can manipulate adults can feel very insecure. A gifted child may try to outsmart parents and teachers. Sophisticated vocabulary and advanced sense of humor can cause gifted children to be misunderstood, which can make them feel inferior and rejected. (This is one reason gifted children prefer to be around older children and adults.)
Perfectionism and Emotional Sensitivities
It's wonderful to have high-level skills, but those skills sometimes create unreasonable expectations. Some gifted children become perfectionists, expecting themselves to get 100 percent scores on every test. Giftedness can also lead to an overactive imagination. These issues can cause behavioral issues.
Perfectionism can lead to fear of failure, in turn causing a gifted child to avoid failure by refusing even to try something (including doing a homework assignment). Keen observation, imagination, and ability to see beyond the obvious can cause a gifted child to appear shy, holding back in new situations in order to consider all the implications.
A gifted child may require full details before answering questions or offering help, making him or her appear socially shy. Intense sensitivity can cause gifted children to take criticism, or even general anger, very personally. Childhood sleights do not roll off their backs.
Sensitivity and a well-developed sense of right and wrong can lead to concern over wars, starving children, pollution, violence, and injustice. If they are overloaded with images and discussions of these issues, they can become introverted and withdrawn or even suffer from "existential depression.”
A Word From Verywell
Virtually all the characteristics of giftedness can make gifted children feel “different,” even at a very early age. It's important, therefore, to get them together with children like them and with people who understand them.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
tonight is Ladies Night! the women of 9Comm are coming together for an epic night of Girl Talk! :-) imagine if you would a slumber party filled with the most dynamic personalities found online today, with pronounced opinions about every topic you can think of. now give each of them a mic and a few games to break the ice and you have history in the making... Guys this is a good chance to get all your questions about how women think and why they think the way they do answered!... And all you ladies out there may want to join in the shenanigans :-) the more the merrier! Call in and share your thoughts ask your questions and have an all around good time
Harshman Travel provides excellent, cost effective deals for all of your traveling needs. Honeymoon & Wedding Packages, Air Fares, Cruises, Vacation Packages, Hotels, Car Rentals, etc. Go to the website http://harshmantravel.com. and join us on The Harshman Travel Radio Show on
http://www.buildingfortunesradio.com/harshman-travel/index.php
This show concentrates on Cruises and what you need to know.
If you want the competitive element use every chance you have to beat the competition.Use all of your marketing pieces competitively.Make sure your e-mail marketing enhances your radio advertising.Make sure your radio advertising enhances your classified ads. Make sure your website enhances your voicemail messages.It's all about integrated marketing principles.
For more information go to: http://www.networkleads.com/imp
Listening often to building fortunes radio onwWW.buildingfortunesradio.com.(386) 445-3585
Aydian Dowling is an FTM transgender activist from Patchogue, Long Island. Aydian documented his own transition on YouTube as a way to voice his thoughts, fears, and successes; as well as offer a support mechanism for others looking to transition. His YouTube channel, A Lions Fears, quickly became popular and to date, has nearly 200 videos, reaches over 11k followers, and has more than 1 million views. In 2011, Aydian was chosen as one of three LGBT persons to have their stories showcased on the Emmy Nominated It Gets Better Project on MTV. Aydian and his now wife, Jenilee, spoke of their soon to be wedding and the challenges transgendered couples face when planning to wed. 2011 also marked the year Aydian Launched Point5cc, the first of its kind, stealth clothing company designed to highlight transgender pride. Point5cc also supply free binders to those in need throughout the community and offers $1,000 scholarships, annually, for individual surgery funds. Point5cc is the first binder exchange program that has been adopted by LGBT centers around the USA and its clothing has been worn by notable advocates such as Dan Savage of the It Gets Better Project, and Buck Angel. Point5cc is also a special feature on the Matthew Shepard Foundation youth website called Matthews Place. Aydian presented to his peers on the topic of Masculinizing Your Body at the 2013 Southern Comfort Conference, in Atlanta, GA; which followed his 2012 presentation on the topic of Emotional Transition. Aydian has also done interviews with Fusion TV, FTM Magazine, the Trevor Project, as well as spoken at colleges such as Montana State and University of Oregon. Aydian continues his mission to help others by answering hundreds of emails a month and documenting his ongoing transition.
On January 22 of this year, Bob Estep was pulled over in a routine traffic stop in Christian County. That's about as far as we go with any consistency between what police say happened and what Estep says happened. New releases from the police department of Sparta this week widen the chasm even more between the reports of how the incident really went down. We'll continue our coverage tonight.
Somehow the Springfield newspaper managed to squeeze the Estep story in amongst the pages crowded with stories of how the "haters" want to encourage discrimination, who the "good Christians" and "bad Christians" are, and their hollow cries to expand the inequalities extended to special classes of people. We'll discuss the upcoming Springfield vote to recall the expansion of special privelege to gay, lesbian, bisexual, and transgendered people.
Join us live at 7PM Missouri Time. Call and participate at 347-677-1835.
Just because I love you one and all this is just for you on the Voice of Joy. Have to be at a wedding anniversary tomorrow. Denny and Linda Mack will be celebrating their 50th . Congratulations. So here is a special for you and Congratulations to them from all of us on the Voice of Joy. Will catch you up on this and much more next Saturday. Have a blessed day and God bless you one and all. Go to church Sunday and pray for all needs.
Speak Life Radio ~ Interviews John West Newest Member of The Hit TV Show Empire!!!!
Blessings family and welcome to this special Broadcast of Speak Life Radio with Special Guest Actor, Producer and Director John West! John has been involved in TV shows and movies such as: Martin, New York Undercover, Southland, Bar-B-Q, The Best Man Wedding, Empire, Snatched, Straight Outta Compton, A Hustler's Life and much more.
Lee Daniels the Producer of Empire has added Actor/Producer John West to the 2nd season of the show. West a native of Chicago is making major moves, will play co/star Taraji P. Henson thugged out boyfriend who has plans to try to get rid of Terence D. Howard from the booing record label. WOW! GET A JUMP ON THE 2ND SEASON OF EMPIRE ... YOU DON'T WANT TO MISS THIS INTERVIEW!!!!!
Tonight features the return of Kergan Edwards-Stout, best selling author, who will join Garrett to talk about all the Gay News that's happening in the LGBT community. Gay Marriage! Healthcare! Gay Mafia! Gay Money! Gay Headlines! You'll hear it all here!
If you've got a hot topic to share, please feel free to join in the conversation!
Meet Adriano Bulla, a servant of Calliope. Born in Italy and lives in London. His publishing journey started in 2005 when his first collection of poetry, Ybo' and Other Lies hit the shelves. He has since published:
a surreal and unusual novel, The Road to London, which mixes poetry and prose (reading at the end of the interview) a collection of short stories, Tales for the Free Mind and Open Heart a second book of poetry, Queer Poems a study on Dante and Coleridge and a grammar book.
His constant search for new and innovative forms in Literature, intense and unusual imagery has been accompanied by a growing awareness of social and psychological issues.
Click to follow Adriano on Twitter.
Click to access all our on-demand shows. Download our shows to enjoy any time, anywhere!
We were joined in Studio tonight by Gervase Peterson of "Survivor Borneo" fame. He told us how he was chosen to be on the show and gave us a complete behind the scenes explination of how it all goes down on set.
Ever thought about crashing a wedding? Well one man accomplished just that. We had Wedding Crasher internet sensation, Joey DiJulio and the groom, Jeff Minetti, call in to tell us their "accidental" cross country friendship story.
Miss Robin gave us an update to a previous update that may help some folks out there looking for employment.
Joe Corrado was in the house trying to make sense of things that just don't.
Hemp? What's that all about? Don't you fret, we explained it in detail for you.
Don't forget to click on the "FOLLOW" button to make sure you never miss an episode & set your ALARM to listen and call into the show LIVE!
| null |
minipile
|
NaturalLanguage
|
mit
| null |
A review of media monitoring services reveals that every cable news network except Fox News has thus far ignored or downplayed the story of the suppressed 2005 photograph of Barack Obama with Nation of Islam leader Louis Farrakhan.
The photograph was released last Saturday by a left-wing journalist. To their credit, the left-wing new media, led by Talking Points Memo, did much of the legwork in reporting further details of the story — such as the fact that the Congressional Black Caucus had invited Farrakhan to meet with them, and that the photograph was deliberately hidden to protect then-Sen. Obama’s nascent presidential campaign.
Still, the networks ignored the story.
A good-faith review of news content via TVeyes, Nexis, and Lexis indicates that the only mainstream news network other than Fox to report the existence of the photograph, and its suppression, was CBS News, which reported it last Friday. (NewsBusters.org argues that even CBS’ coverage danced around connections between Democrats, Farrakhan, and the suppression effort.)
The most recent coverage of the issue was on Fox & Friends on Saturday, Jan. 27, when the noted defense lawyer, liberal Democrat, and Obama supporter Alan Dershowitz declared that he would not have campaigned for Obama in 2008 had he known of Obama’s association with “virulent anti-Semite and anti-American” Farrakhan.
In contrast, cable news networks focused obsessively in 2016 on President Donald Trump’s supposed reluctance to disavow support from Ku Klux Klan leader David Duke, even though the two had no prior association and Trump had repeatedly denounced him in the past.
Jake Tapper of CNN pressed Trump during an interview in February 2016 on his supposed refusal to disavow Duke — even though Trump had condemned Duke only two days before.
That became the pretext for claims by the media and Democrats that Trump tacitly supported white supremacists — claims that have persisted to this day, though evidence tying Obama to black supremacists — including his pastor of some two decades, Jeremiah Wright — is far stronger.
As freelance writer Jeryl Bier noted in the Wall Street Journal: “If Republican lawmakers were holding strategy sessions with Mr. Duke, their party would rightly be held to account. Why shouldn’t the Democrats and the Congressional Black Caucus be held to the same standard?”
Joel B. Pollak is Senior Editor-at-Large at Breitbart News. He was named to Forward’s 50 “most influential” Jews in 2017. He is the co-author of How Trump Won: The Inside Story of a Revolution, is available from Regnery. Follow him on Twitter at @joelpollak.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
AT&T Plaza
AT&T Plaza (formerly Ameritech Plaza and SBC Plaza) is a public space that hosts the Cloud Gate sculpture. It is located in Millennium Park, which is a park built to celebrate the third millennium and which is located within the Loop community area of Chicago, Illinois in the United States. The sculpture and the plaza are sometimes jointly referred to as Cloud Gate on the AT&T Plaza.
It was opened in the summer of 2004 with the initial unveiling of the sculpture during the grand opening weekend of the park. Ameritech Corporation/SBC Communications Inc. donated US$3 million for the naming right to the space. The plaza has become a place to view the McCormick Tribune Plaza & Ice Rink and during the Christmas holiday season, the Plaza hosts Christmas caroling.
Details
Lying between Lake Michigan to the east and the Loop to the west, Grant Park has been Chicago's front yard since the mid-19th century. Its northwest corner, north of Monroe Street and the Art Institute, east of Michigan Avenue, south of Randolph Street, and west of Columbus Drive, had been Illinois Central rail yards and parking lots until 1997, when it was made available for development by the city as Millennium Park. Today, Millennium Park trails only Navy Pier as a Chicago tourist attraction.
The plaza is located above Park Grill, above and behind the McCormick Tribune Plaza & Ice Rink, adjacent to the Chase Promenade, and between the North and South Boeing Galleries. The plaza and sculpture sit atop the 300-seat $6 million Park Grill, which opened in November 2003 behind the McCormick Tribune Plaza & Ice Rink. The surface of the plaza is concrete. The plaza is composed of of concrete pavers. Each paver is , and each is thick.
History
The plaza was originally named Ameritech Plaza for Ameritech Corporation, the corporate sponsor, who donated $3 million for the sculpture-hosting plaza's naming rights. By the time the park officially opened in 2004, Ameritech had merged with SBC Communications and the plaza was called SBC Plaza. When SBC acquired AT&T and subsequently changed the name from SBC to AT&T in 2005, the name of the plaza changed again.
Cloud Gate was originally estimated to weigh because it was impossible to estimate the thickness of the steel compatible with the desired aesthetics. The final piece, however, weighs and care had to be taken in supporting it. The roof of the Park Grill, upon which Cloud Gate sits, had to be strong enough to bear the weight. A large retaining wall separating Chicago's Metra train tracks from the North Grant Park garage travels along the back side of the restaurant and supports much of the sculpture's weight. This wall, along with the rest of the garage's foundation, required additional bracing before the piece was erected. In June 2004, when construction of the shell began, a large tent (pictured left) was erected around the piece in order to shield it from public view.
Activities
In 2006, annual Christmas caroling began at the plaza. Following Thanksgiving, weekly sing-alongs are led by choral groups including Bella Voce, Chicago Mass Choir, and Chicago Children's Choir.
Because of its elevation above the McCormick Tribune Plaza & Ice Rink, the plaza has become a prime viewing location for jazz concerts held during the summer at the McCormick Tribune Plaza. McCormick Tribune Plaza is located below and to the west of AT&T Plaza as well as adjacent to Michigan Avenue's Historic Michigan Boulevard District, which are slightly further west.
Notes
References
External links
City of Chicago Millennium Park
Millennium Park map
City of Chicago Loop Community Map
Category:2004 establishments in Illinois
Category:Buildings and structures celebrating the third millennium
Category:Millennium Park
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Ultrasonic studies on lamivudine: beta-cyclodextrin and polymer inclusion complexes.
The aim of the present study is to enhance the solubility and stability of drugs in addition of water-soluble polymer and carbohydrate complexes. The data show that the polymer polyvinyl alcohol (PVA) interacts with the free Lamivudine and with the Lamivudine; beta-cyclodextrin (beta-CD) inclusion complex, in both cases with particular intermolecular interaction was studied using ultrasonic technique under different concentrations at a temperature 303 K. Consequently, the reason of this study was to improve the biological performance of the drug through enhancing its solubility and stability. The binary and ternary mixtures prepared inclusion complexes of Lamivudine in beta-CD and PVA. The presence of PVA, changes the drug: beta-CD interaction, a Lamivudine: beta-CD: PVA complex was formed. In addition, the presence of PVA produces a strong increase in the binding constant at a particular concentration (1.25%). In the ternary complex, the Lamivudine is wrapped at both ends for the beta-CD. In this complex, the polymer seems to act as a bridge between both beta-CD molecules that bind the Lamivudine.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
The interrelationship of somatostatin and guanylate cyclase activity.
Somatostatin has been shown to inhibit the release of various polypeptide hormones including insulin, glucagon, gastrin, thyroid stimulating hormone, and growth hormone. The mechanism by which somatostatin inhibits the release of these various polypeptide hormones has not been fully elucidated. It has been reported that somatostatin increases the level of the second messenger cyclic GMP in rat brain and in the anterior pituitary gland. The present investigation was designed to determine if these responses seen in the anterior pituitary gland and brain were due to activation of guanylate cyclase [GTP-pyrophosphate lyase (cyclizing), E.C.4.6.1.2.], the enzyme that catalyzes the formation of cyclic GMP. Somatostatin at a concentration of 2 pM enhanced guanylate cyclase activity two-fold in rat cerebrum and anterior pituitary gland. This enhancement of guanylate cyclase activity was also seen in rat liver, pancreas, stomach, and small intestine at the same concentration of somatostatin. Increasing the concentration of somatostatin to 20 microM, caused a marked inhibition of guanylate cyclase activity in all these tissues. Dose-reponse curves done on gastric guanylate cyclase activity revealed that over a concentration range of 2 pM to 0.2 microM, somatostatin had a stimulatory effect on guanylate cyclase activity while at concentrations above 10 microM somatostatin was inhibitory to guanylate cyclase activity. The biphasic pattern of enhancement of guanylate cyclase activity at lower concentrations of somatostatin and inhibition at higher concentrations may help to explain some of the discrepancies seen with previous investigations with somatostatin, hormone release, and cyclic nucleotide metabolism.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Effect of superficial harrowing on surface properties of sand with rubber and waxed-sand with fibre riding arena surfaces: a preliminary study.
A recent epidemiological study identified various aspects of arena surfaces and arena surface maintenance that were related to risk of injury in horses and that arena maintenance is important in reducing injury risk. However, there has been little research into how properties of arena surfaces change with harrowing. This study aimed to compare the properties of different arena surface types pre- and post-harrowing. The Orono Biomechanical Surface Tester fitted with accelerometers and a single- and a three-axis load cell was used to test 11 arenas with two different surfaces types, sand with rubber (SR) and waxed-sand with fibre (WSF). Three drop tests were carried out at 10 standardised locations on each arena. Mixed models were created to assess the effect of surface type, pre- or post-harrowing, and drop number on the properties of the surface, including maximum horizontal deceleration, maximum vertical deceleration, maximum vertical load and maximum horizontal load. Post-harrowing, none of the parameters were altered significantly on SR. On WSF, maximum vertical deceleration and maximum vertical load significantly decreased post-harrowing. The differences in the effects of superficial harrowing on SR and WSF could be attributed to the different compositions and sizes of the surface material. The results suggest that different maintenance techniques may be more suitable for different surface types and that the effects of superficial harrowing are short-lived due to the rapid re-compaction of the surface with repeated drops on WSF. Further work is required to determine the effects of other maintenance techniques, and on other surface types.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
[Anxiety disorders and unexplained syncopes of presumed vaso-vagal origin].
The impact of anxiety neurosis on the diagnosis and treatment of patients with unexplained syncope (S) was assessed in 178 patients (91 women and 87 men) with an average age of 36.5 +/- 20 years, presenting with 10.7 +/- 24 episodes of S). None had evidence of underlying cardiac disease apart from 7 patients with mild hypertension. All patients underwent a tilt test (TT) at 60 degrees for 45 minutes. A bolus of isoproterenol was injected intravenously in subjects with negative TT. After the test, the patients were classified according to the presence (n = 38) or absence (n = 140) of anxiety neurosis based on the DSM III-R diagnostic. The TT was positive in 76 patients, 9 of whom had a cardioinhibitory reaction with prolonged asystole. Patients with anxiety had more episodes of S (24 +/- 43 versus 7 +/- 13; p = 0.001), a shorter interval between S (11.5 +/- 23 months versus 12.5 +/- 20 months, p = 0.02) but more negative TT (27/38 versus 75/140; p = 0.05). One hundred and sixty-eight patients were followed up : 10 were lost to follow-up. Preventive treatment was undertaken in 59 patients who were representative of the whole group with respect to age (30 +/- 18 years 39 +/- 21 years : p = 0.004). After an average follow-up of 24.5 +/- 15 months, 26 patients (15%) experienced a recurrence of S. The recurrence rate was identical in patients with positive and negative TT and in treated and untreated cases. On the other hand, recurrence was higher in those with anxiety (12/25 versus 14/117; p = 0.001) who also had less improvement of symptoms (12/15 versus 74/120; p = 0.001). The "anxiety" variable was therefore identified as being the only predictive factor for recurrence of syncope. The authors conclude that in patients referred for investigation of unexplained syncope, some suffer from anxiety neurosis, in whom the TT is usually negative, and have a higher risk of recurrence. They justify a specific therapeutic management.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
+----------+-------------------+
| Dactyl | R Documentation |
+----------+-------------------+
Edgeworth's counts of dactyls in Virgil's Aeneid
------------------------------------------------
Description
~~~~~~~~~~~
Edgeworth (1885) took the first 75 lines in Book XI of Virgil's *Aeneid*
and classified each of the first four "feet" of the line as a dactyl
(one long syllable followed by two short ones) or not.
Grouping the lines in blocks of five gave a 4 x 25 table of counts,
represented here as a data frame with ordered factors, ``Foot`` and
``Lines``. Edgeworth used this table in what was among the first
examples of analysis of variance applied to a two-way classification.
Usage
~~~~~
::
data(Dactyl)
Format
~~~~~~
A data frame with 60 observations on the following 3 variables.
``Foot``
an ordered factor with levels ``1`` < ``2`` < ``3`` < ``4``
``Lines``
an ordered factor with levels ``1:5`` < ``6:10`` < ``11:15`` <
``16:20`` < ``21:25`` < ``26:30`` < ``31:35`` < ``36:40`` <
``41:45`` < ``46:50`` < ``51:55`` < ``56:60`` < ``61:65`` <
``66:70`` < ``71:75``
``count``
number of dactyls
Source
~~~~~~
Stigler, S. (1999) *Statistics on the Table* Cambridge, MA: Harvard
University Press, table 5.1.
References
~~~~~~~~~~
Edgeworth, F. Y. (1885). On methods of ascertaining variations in the
rate of births, deaths and marriages. *Journal of the [Royal]
Statistical Society*, 48, 628-649.
Examples
~~~~~~~~
::
data(Dactyl)
# display the basic table
xtabs(count ~ Foot+Lines, data=Dactyl)
# simple two-way anova
anova(dact.lm <- lm(count ~ Foot+Lines, data=Dactyl))
# plot the lm-quartet
op <- par(mfrow=c(2,2))
plot(dact.lm)
par(op)
# show table as a simple mosaicplot
mosaicplot(xtabs(count ~ Foot+Lines, data=Dactyl), shade=TRUE)
| null |
minipile
|
NaturalLanguage
|
mit
| null |
At its “Never Is Now” anti-Semitism summit this week, the Anti-Defamation League gave its ADL Americanism Award to Lt. Gen. Jay Silveria, superintendent of the Air Force Academy, for his famed speech last September denouncing white racists for scrawling antiblack slurs on the dorm door of a black student.
Silveria’s order to racists to “Get out” (an admiring reference to the antiwhite horror movie) was celebrated by the media, with Silveria being declared 2020 presidential timber and an admirable rival to the deplorable Donald Trump. Silveria was especially adulated for dog-whistling:
We would also be tone-deaf not to think about the backdrop of what’s going on in our country—things like Charlottesville and Ferguson, the protest in the NFL.
The only problem with the ADL’s ceremony honoring Lt. Gen. Silveria’s “commitment, dedication, and leadership in the fight against hate” was that six days earlier the Air Force Academy had admitted that the graffiti had been a hoax perpetrated by an unnamed black student.
In other words, the Air Force Academy incident had indeed been a racist hate crime, but, contra General Silveria, one perpetrated to spread fear and loathing against its white victims.
The ADL, however, is not in the business of letting reality get in the way of business.
“The ADL is not in the business of letting reality get in the way of business.”
Granted, the ADL, with $81 million in net assets, isn’t as rich as the money-crazed Southern Poverty Law Center, America’s most lucrative hate organization. Still, the ADL has done very well for itself in its 104 years of activity, collecting, for instance, $56,000 over the years from Donald Trump.
Trump’s election has provided the ADL with a profitable pretext for ramping up fund-raising efforts. Jewish Insider reported a week after the 2016 election:
The Anti-Defamation League has seen a dramatic increase in online donations since the election of Donald Trump, new data shared with ‘Jewish Insider’ showed…. Last Wednesday, the day after the surprising election results, the ADL received a 50X increase in online donations, mostly from new donors…. “We are heartened by the outpouring of support that we and reportedly other nonprofit organizations in this field have received over the past week,” ADL CEO Jonathan Greenblatt told ‘Jewish Insider’.... ADL is kicking off its inaugural “Never is Now!” two-day summit in New York City on Thursday to address some of the most urgent challenges facing the Jewish community and identify new strategies to effectively combat anti-Semitism. Tickets to the events are sold out, according to ADL.
Last winter, for example, the ADL had been doing well off blaming the sudden spate of threats against Jewish community centers upon Trump.
But then it turned out that eight of the calls had been made by a black leftist journalist named Juan Thompson. And next it emerged that up to 2,000 threats were the work of an energetic young Jewish man in Israel with dual Israeli-American citizenship, Michael Ron David Kadar.
Undeterred, the ADL responded to Kadar’s arrest in Israel not by demanding his extradition to the U.S. but by issuing an amusing press release announcing:
While the details of this crime remain unclear, the impact of this individual’s actions is crystal clear: These were acts of anti-Semitism.
Similarly, Silveria’s disgrace didn’t dissuade Jonathan Greenblatt, the head of the ADL, from honoring him for stoking hatred.
Similarly, Kansas State University discovered last week that its nationally publicized antiblack hate crime on Halloween was actually another antiwhite hate hoax. Likewise, the destruction of a Jewish ceremonial tent at Kansas State earlier in the fall had not been anti-Semitic violence by vicious Trump supporters as originally theorized, but an Act of God: A gust of wind had blown over the tent.
Nonetheless, KSU president Richard Myers announced, without mentioning the hoax, that he was suspending classes for two hours on Tuesday of this week for a rally against racism. With 24,766 students and an out-of-state tuition of $23,000, wasting a quarter of a school day at KSU to fulfill the goals of the hoax would appear to cost nearly a million dollars.
That would seem to be atrocious management. Yet ask yourself: What bureaucrat has ever gotten in trouble for falling for a hate hoax? After all, the concept of a “hate hoax” has been successfully kept out of the mainstream mentality. In all its years since its founding in 1851, The New York Times has never once used the words “hate” and “hoax” consecutively.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Mediterranean Italian Spanish Tuscan Homes & Decor Ideas.
Bathroom Decor Print available in two sizes, and Printed on quality matte photo paper and shipped anywhere in the US. For inexpensive funny home decor to fill your empty wall spaces click through to the store and see all the options offered.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Rogers Media uses cookies for personalization, to customize its online advertisements, and for other purposes. Learn more or change your cookie preferences. Rogers Media supports the Digital Advertising Alliance principles. By continuing to use our service, you agree to our use of cookies.
We use cookies (why?) You can change cookie preferences. Continued site use signifies consent.
UPDATED: Don't worry, the number-crunchers over at Finance are much better at this stuff. - Macleans.ca
UPDATED: Don’t worry, the number-crunchers over at Finance are much better at this stuff.
With Jim Flaherty’s surprise mini-budget-like-thing mere moments from being unveiled in Victoria, ITQ would just like to reassure readers that she’s almost positive that his numbers will hold up better under the scalpel-like scrutiny of Kevin Page, Parliamentary Budget Officer At Large than those put forward by certain cabinet colleagues:
Parliament’s budget watchdog has put a price tag of under $1.2 billion on the Liberal’s employment insurance proposal, showing the Conservative government “wildly overestimated” the cost at $4 billion for partisan purposes, Montreal Liberal MP Marlene Jennings said Wednesday.
The government rejects her accusation and stands by its $4 billion estimate as a “prudent” one that includes the impact on the labour market of making benefits easier to get, said Ryan Sparrow, a spokesman for Human Resources Minister Diane Finley.
Jennings told Canwest News Service a report prepared for the Liberals by parliamentary budget officer Kevin Page contains an estimated cost of less than $1.2 billion for the Grit proposal for a national eligibility standard for employment insurance benefits. That is less than the $1.5 billion the Liberals estimated themselves.
The finding shows the Conservative government distorted the price during bipartisan talks that broke down over the summer, Jennings said in a telephone interview from Montreal. A senior government official had put the $4 billion price tag on the proposal as the possibility faded of a consensus by the committee, of which Jennings is a member. […]
You know what would be helpful? If both sides would actually release their respective reports, rather than just battling it out in the press using distinctlyunhelpful phrases like “prudent” and “wildly overestimated”. Just show us the numbers, y’all. We can make up our own minds as to which numbers seem more reflective of reality.
Almost Done!
Please confirm the information below before signing up.
{* #socialRegistrationForm *}
{* socialRegistration_firstName *}
{* socialRegistration_lastName *}
{* socialRegistration_emailAddress *}
{* socialRegistration_displayName *}
By clicking "Create Account", I confirm that I have read and understood each of the website terms of service and privacy policy and that I agree to be bound by them.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Staphylococcal scalded skin syndrome I. Purification of exfoliatin and maternal transmission of neutralizing ability against exfoliatin.
Exfoliatin, an exotoxin produced by phage group 2 staphylococci, was separated and was analyzed. The toxin was separated in a single peak by Sephadex G-100 gel filtration and subsequent DEAE Sephadex chromatography. By acrylamide gel disc electrophoresis this single peak was found to contain 2 components both of which posessed exfoliative activity and the same antigenic properties. Their molecular weight, however, differed; the major component had a molecular weight of 30,000 and the minor component had 27,000. The toxin was stable when heated at 60 degrees C for 40 minutes, and labile at 100 degrees C for 20 minutes. The toxin was immunogenic to mice and produced antibody which could be transmitted to neonatal mice to neutralize exfoliative activity of the toxin. Innoculation of the bacteria caused the production of antibodies against both exfoliatin and alpha-hemolysin in mice. The antibodies were also transmitted to neonatal mice to neutralize the exfoliative activity of exfoliatin.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Doncaster could be set to welcome up to 10 families fleeing persecution in Syria, the local authority has revealed today.
This comes after the council agreed to ‘work with the Government’ to help it fulfil its pledge of accepting 20,000 Syrian refugees over the next five years.
The council says it will not be able to make a decision on the number of Syrian refugees Doncaster will support until the Home Office provides full detail of their programme.
However, a spokesperson confirmed that early indications are that up to 10 families could be accepted into the borough.
Mayor of Doncaster Ros Jones said: “We have said we would help with this refugee crisis and are working closely with our colleagues across the region to provide appropriate support. We are still waiting for the detail of the Government’s resettlement programme and until this is forthcoming there is not much more we can add.
“The reality is we are talking about housing relatively small numbers of refugees in the private sector which will not be a burden on council tax payers.”
Councils across the country have also signed up to accept dispossessed families, and Chancellor of the Exchequer, George Osborne, confirmed that the money to house Syrian refugees will come from the UK’s £12 billion international aid budget.
When polled earlier this month, the overwhelming majority of Star readers said they did not think Doncaster should open its doors to refugees.
A total of 72.1 per cent of people said ‘no’ when asked if Doncaster should open its doors to refugees.
Some 127 people or 27.9 per cent of those polled said they thought the borough should welcome refugees.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Roman Camp Country House Hotel, Callander
0
Your rating: None
The four star luxury Roman Camp Country House Hotel provides beautifully appointed suites and guest rooms, an elegant award winning restaurant, along with spacious private rooms for functions and events. Our country house is located within easy travel of Glasgow, Edinburgh and Stirling, providing a secluded comforting retreat from city life.
The Roman Camp Country House Hotel can be found in the Loch Lomond & Trossachs National Park, the first national park to be created in Scotland. The hotel is located within the boundaries of the scenic village of Callander, but secluded in private gardens and parkland in the area more affectionately referred to as 'Braveheart Country' - only 15 miles west of Scotland’s historic and newest city, Stirling.
Visiting Scotland's main cities, including Glasgow, Perth and Edinburgh is easy as Callander is centrally located. These main cities can be reached within an hour, by using the main routes or you can meander along the scenic back roads, where the sight of Scotland's wildlife is possible and a chance to see the hidden views of our country.
About Roman Camp Country House Hotel, Callander
Rooms at Roman Camp Hotel
The Roman Camp Hotel has a wide selection of bedrooms, dining rooms, and lounges to hide away in cosy corners, allowing you to feel at home and enhancing your stay and the grand features of the hotel.
Bedrooms
A room in which to start the day or retreat to as evening turns to night - sumptuous, relaxed and pampered. Each of our bedrooms has a distinctive style and character of its own and all are individually designed, furnished and decorated with William Morris wall coverings to ensure your stay is as comfortable as possible. Bedrooms are fitted with the latest technology to make your stay as pleasant as possible, including en-suite facilities, telephones, LCD flat-screen TVs with satellite, radio and much much more.
Dining rooms and Restaurants
Our main restaurant provides a fine dining experience along with an elegant atmosphere in an inspiring oval room of elegant proportions, softly decorated in a modern, classic style. By evening, with candle-lit tables laid with crystal and silverware, magic transcends the room. For a more intimate or private dining, we have other smaller rooms, which can be set for special occasions.
Relax and Rest
Our public rooms are designed to allow you to relax after a hard day sightseeing or to unwind after a day at a conference. During winter, our open real fires bring a sense of warmth and atmosphere especially when the gardens are covered in untouched, white, crisp snow.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Blood rheology in deep venous thrombosis--relation to persistent and transient risk factors.
The risk of recurrence in patients with symptomatic deep venous thrombosis (DVT) is higher in patients with persistent risk factors than in patients with transient risk factors. The purpose of this prospective study was to investigate the course of rheological variables in patients with DVT up to 1 year after the acute event in relation to risk factors. In 37 patients with proven DVT plasma fibrinogen, plasma viscosity, red cell aggregation, whole blood viscosity, hematocrit and platelet aggregation were studied in the acute phase, at 6 weeks and at 12 months. In the acute phase, patients had higher fibrinogen (medians and ranges; 450 [270-611] vs. 247 [170-340] mg/dl, p < 0.01), plasma viscosity (1.67 [1.48-1.96] vs. 1.60 [1.50-1.70] mPa s, p < 0.01), red cell aggregation (9.76 [5.87-12.66] vs. 5.66 [3.67-8.46] arbitrary units; p < 0.01) and whole blood viscosity (5.78 [5.61-5.87] vs. 5.59 [5.27-5.9] mPa s, p < 0.01), but lower hematocrit (40 [32-46] vs. 45 [38-50]%, p < 0.01) and platelet aggregation (by epinephrine: 41 [13-85] vs. 79 [29-91]%, (p < 0.01) than controls. During the 1-year follow-up, fibrinogen, plasma viscosity, red cell aggregation and whole blood viscosity constantly decreased, whereas hematocrit and platelet aggregation increased in the total of patients (all p < 0.01). Subgroup analysis according to risk factors showed that at 12 months patients with persistent risk factors (N = 21) had higher plasma fibrinogen (357 [235-450] vs. 247 [214-335] mg/dl, p < 0.01), plasma viscosity (1.65 [1.50-1.80] vs. 1.59 [1.42-1.77] mPa s, p < 0.05) and red cell aggregation (7.82 [6.0-11.3] vs. 6.3 [5.2-7.1] arb. units, p < 0.01) than patients with transient risk factors (N = 16). Compared with controls, these variables were increased in patients with persistent risk factors (all p < 0.01), but not in patients with transient risk factors (all n.s.). In patients with persistent risk factors rheological changes are still present 1 year after acute DVT, whereas in patients with transient risk factors blood rheology returns to normal. Further studies are needed to clarify whether blood rheology might be helpful to identify patients at high risk of recurrence.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
In a Federal Register notice published on August 6, 2014 the U.S. Department of Commerce renewed the existing anti-dumping and countervailing duty orders against imports of polyethylene terephthalate film, sheet and strip from India and Taiwan. The original order was imposed on July 1, 2002, with the renewed order thus exceeding the five-year 'sunset' standard that generally applies to trade-remedy measures under WTO agreements (i.e., article 21.3 of the Agreement on Subsidies and Countervailing Measures and article 11.3 of the Anti-dumping Agreement).
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Dave and I have come to the conclusion that winter vacations in snowy mountain retreats are the ultimate.
In short–Banff is so stunningly beautiful I can’t imagine anyone not wanting to visit it.
I don’t know if I could adequately put in words how lovely of a time we had up there. So I will show you pictures instead.
(Above + below are views from the terrace of Banff Springs Hotel, where in summer time you can have a meal)
We spent time at Lake Louise on New Years Eve – and even went on a sleigh ride around the lake! It was so charming.
(Later that evening Dave and I had planned out being out in town. But when it came right down to it, all the places we wanted to be had a cover charge…and it was too cold to loiter on the streets till midnight. So…we bought a bottle of wine and spent New Years in our hotel room…watching a Mythbusters marathon. Huzzah!)
(it was SO cold that day)
New Years Day was relaxed. We went for a drive intending on visiting a Ghost Town, but the road was closed from snow. So we kept going down some other back roads and discovered some beautiful locations where no one else was around.
Lake Minewanka
Two Jack Lake–Totally frozen over.
(These boots are older than me. They have lived through winters in Northern BC. They are better than my new winter boots that I got only last year. My mom is not getting these boots back.)
The roads were not as treacherous as I thought they would be. The only time we were delayed was going into Banff (they closed the road between Revelstoke + Golden–we had to wait in Revelstoke for 3 hours while they cleaned up the accident)
While the roads were not as treacherous as I thought they would be, it was still too dangerous to just stop wherever to take pictures. And many of the viewpoints have not been cleared of the snow so there was no way to get to them. Therefore…all the following images were taken while I hung out of the window of our car. That’s safe, right?
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Thousands of children roam the streets of Manila, forced to scavenge, beg or even resort to prostitution because of poverty. For many of them, education is not an option as their families struggle to put food on the table, much more send their kids to school. Dropouts are also at an alarming rate, further dampening the possibility of their breaking the poverty cycle.
Having been assigned to the Philippines, French Jesuit priest, Fr. Pierre Tritz learned about the plight of Filipino street children and embarked on a mission to set it right. He gave up his French citizenship to become a naturalized Filipino citizen, and established the Educational Research and Development Assistance (ERDA) Foundation in 1974 to give street children a fighting chance at a better life.
During a visit to the Philippines, Ramon and Dellie Sierra had a chance encounter with Fr. Pierre, who, as is typical of him to everyone he meets, immediately filled them in on the realities Filipino street children face and how ERDA aims to improve them. Inspired and motivated when they came back to Spain, the couple founded the Intramuros Pro Infancia Filipina in 2003 to help ERDA with its funding needs.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Q:
SQLite count, group and order by count
I have a table that looks like this:
FOO BAR BAZ
----+----+----
foo1 bar1 baz1
foo2 bar3 baz2
foo3 bar1 baz3
foo4 bar1 baz4
foo5 bar3 baz5
foo6 bar1 baz6
foo7 bar2 baz7
And as a result I would like to get the count of how many times each bar appeared in the table.. so, the output I'm looking for looks like this:
BAR COUNT
-----+-----
bar1 4
bar3 2
bar2 1
Can I do a query for something like this in SQLite? I guess it should be pretty easy, but I am not an SQL programmer by any means, and I just need this simple query as a part of a python script.. Thank you.
A:
SELECT foo, count(bar) FROM mytable GROUP BY bar ORDER BY count(bar) DESC;
The group by statement tells aggregate functions to group the result set by a column. In this case "group by bar" says count the number of fields in the bar column, grouped by the different "types" of bar.
A better explanation is here: http://www.w3schools.com/sql/sql_groupby.asp
| null |
minipile
|
NaturalLanguage
|
mit
| null |
About
Have you ever felt like you are living in a time that you do not quite understand? As the days go on, the world around us is changing at such a rapid pace that soon it will become unrecognizable when looking back on past times in our lives. However, because of the redemption work of Jesus Christ on the cross almost 2,000 years ago, we know that there is a greater country waiting for us, the coming Kingdom of Heaven! And the greatest thing of all is this; the God that we serve NEVER CHANGES! He is the same yesterday, today, and forever! He also sent His Holy Spirit to be our Counselor in the midst of the trials of this life! How amazing is that?!
This blog will seek to bring about biblical encouragement to you (my brothers and sisters in Christ) and infuse as much faith, hope, and love I possibly can regarding the state of world affairs by the power of God’s Holy Spirit!
I strongly believe the Lord has been moving in powerful ways and is about to start working in ways we can’t even begin to fathom… my heart’s desire is for all of us to be ready to do His work here and now to prepare for His Coming Kingdom!
Rejoice in the Lord always. I will say it again: Rejoice!Let your gentleness be evident to all. The Lord is near.Do not be anxious about anything, but in every situation, by prayer and petition, with thanksgiving, present your requests to God.And the peace of God, which transcends all understanding, will guard your hearts and your minds in Christ Jesus.
Finally, brothers and sisters, whatever is true, whatever is noble, whatever is right, whatever is pure, whatever is lovely, whatever is admirable—if anything is excellent or praiseworthy—think about such things.Whatever you have learned or received or heard from me, or seen in me—put it into practice. And the God of peace will be with you. ~Philippians 4:4-9 (NIV)
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Archive for July 2017
Kids love to create and make new things. This is why you as a parent often be surprised that your cute little kid has made some cool things with things found around the house. First of all, this is unsafe. Kids becoming more curious and they discover how fun it is to create things out of scratch. If you don’t give them toys that can enable them to do that, they will gather the materials out of things that they will find lying around the house. There’s always the risk that they will grab unsafe things that can harm them. This is where creative play toys can help you. You can buy these toys from reputed stores that are known for selling good at educational toys. With them, you’re giving your kid an outlet so he can unleash his creative side. If you’re not going to give him the right tools and equipment, you’re just wasting a huge opportunity to develop your kid. If he’s able to create cool things out of nothing, can you imagine what he can do with the right tools and equipment? In a way, you’ll be stunting his growth and he won’t be able to reach his full potential if you don’t give him the right tools and equipment.
Choose right toys for boys and girls
There are a lot of great creative play toys in the educational toys online store right now and your kid can take advantage of them. For example, your little boy can play with toy bricks, toy links, wooden blocks and the likes. They’re very simple toys but they’re highly popular for a reason. With these toys, your little boy will be building structures and that’s a good outlet for him. What if you have a little girl? You don’t have to worry because there are a lot of creative play toys for her as well. To know more about girl’s toys read more about educational toys. A toy sewing machine is a very popular choice and that can surely unleash your little girl’s creative side. Additionally, there are also other toys like pottery wheels and clay tools that your little girl can mold into great things. There are a lot of creative play toys that boys and girls can enjoy. An activity center is another popular option because it offers a lot of activities. When buying toys, it’s a good idea to buy something that offers a lot in terms of fun and entertainment. This way, your kid will be playing for hours playing different games.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
CASABLANCA, Morocco — More than 10,000 Moroccans chanting “death to Israel” took to the streets of Casablanca on Sunday to protest the United States’ decision to move its embassy in Israel to Jerusalem.
The marchers carried Palestinian flags and placards that called Jerusalem “Palestine’s eternal capital.” Most appeared to be Islamists, with women wearing head scarves and marching separately from the men.
The United States opened its embassy in Jerusalem on May 14, relocating it from Tel Aviv — a move that reversed decades of United States policy, delighted Israel and infuriated Palestinians.
The status of Jerusalem is one of the thorniest obstacles to forging a peace deal between Israel and the Palestinians, who with broad international backing want East Jerusalem, captured by Israel in the 1967 Middle East war, as their capital.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Urinary Incontinence Support Group
Urinary incontinence is the involuntary excretion of urine from one's body. It is often temporary, and it almost always results from an underlying medical condition. Stress, functional, and urge incontinence are the most common forms. Share with other support group members and learn from their stories and experiences.
bladder cancer
For the last 3 months I have been passing blood and blood clots in my urine. My doctor sent me to a urilogical specislst. There they did a CT scan and later preformrd a cyscopathy??? any way the cyscopathy showed me the cancer in my bladder. I am having surgery on the 22nd to have it removed.
I am so sorry they found cancer, but I am very glad you opted for removal. I have two family members that didn't, and they're cancer came back and has continued to come back. My grandfather had bladder cancer and opted for removal. He never had a problem again. I have had tumors removed, but they've always been benign. If they ever turn into anything more then I'll have a removal too. I'll be praying for you as you go through all of this. Good luck!
Dave,
I am new to this group as of today...but I wanted to give you my support during your chemo and follow up. I too have been passing blood in my urine for a while now....and my incontinence has been getting progressively worse.
The urologist has not said cancer, but has suggested a suprapubic catheter. I have suspected that I may have cancer or serious infection of the bladder which will warrant surgery.
Anyway, I just wanted to say, Thanks for posting and good luck...
Well it has turned out to be invasive blAadder cancer. I will have 3 months of chemotherapy first. Then they are going to remove my bladder and replace it with a neobladder. Then I will have 6 months of chemo again. After all of this only God knows what happens next.
A friend sent this to me..As far as I can see, grief will never truly end.It may become softer overtime, more gentleand some days will feel sharp.But grief will last as long as Love does - ForeverIt's simply the way the absence of your loved onemanifests in your heart. A deep longing accompaniedby the deepest Love some days. The heavy fog mayreturn and the next day, it may recede.Once again, it's...
theatre and I are there already. I'm having a very berry tea with crackers, cheese and cherry tomatoes and she's having a joint with some beer and we're both on really comfy recliners on thick pile carpet. we need some help with the decor if anyone is around??
All content posted on this site is the responsibility of the party posting such content.
Participation on this site by a party does not imply endorsement of any other party's content,
products, or services. Content should not be used for medical advice, diagnosis, or treatment.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Regulation of Zap-70 by Src family tyrosine protein kinases in an antigen-specific T-cell line.
To further understand the interactions between Zap-70, Src family kinases, and other T-cell proteins, we have examined the regulation of Zap-70 in the antigen-specific T-cell line BI-141. By analyzing derivatives containing an activated version of either p56lck or p59fynT, it was observed that the two Src-related enzymes augmented T-cell receptor (TCR)-mediated tyrosine phosphorylation of Zap-70, as well as its association with components of the antigen receptor complex. Importantly, the accumulation of TCR.Zap-70 complexes quantitatively and temporally correlated with the induction of tyrosine phosphorylation of the CD3 and zeta chains of TCR. Using a CD4-positive variant of BI-141, we also found that the ability of Zap-70 to undergo tyrosine phosphorylation and associate with TCR was enhanced by aggregation of TCR with the CD4 co-receptor. Further studies allowed the identification of two distinct pools of tyrosine-phosphorylated Zap-70 in activated T-cells. While one population was associated with TCR, the other was co-immunoprecipitated with a 120-kDa tyrosine-phosphorylated protein of unknown identity. In addition to supporting the notion that Src-related enzymes regulate the recruitment of Zap-70 in TCR signaling, these data added further complexity to previous models of regulation of Zap-70. Furthermore, they suggested that p120 may be an effector and/or a regulator of Zap-70 in activated T-lymphocytes.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
News Release » 
North Carolina state park and AmeriCorps partner to improve longleaf pine forest
RALEIGH -- A crew of AmeriCorps volunteers is at Singletary Lake State Park this week planting 13,000 wiregrass plants to improve the park’s longleaf pine forest and its habitat for the endangered red-cockaded woodpecker, according to the N.C. Division of Parks and Recreation.
The partnership also involves Quail Unlimited, which provided a $2,500 grant set aside to nurture warm-season grasses in areas off-limits to hunting. The wiregrass plants originated at the North Carolina Forest Service’s Claridge Nursery in Goldsboro.
Singletary Lake State Park in Bladen County has two known red-cockaded woodpecker colonies. The species prefers an “open” longleaf pine forest with few mid-story trees such as oak. A longleaf pine ecosystem is best supported by a wiregrass ground cover and regular prescribed burns. Establishing wiregrass is a labor-intensive process of planting by hand, and the team has also removed small hardwoods from the state park’s longleaf pine forest.
“The partnership with AmeriCorps has been a valuable experience for the state parks system, increasing our efficiency and giving us more options in managing the parks’ natural resources,” said Lewis Ledford, state parks director.
This is the third year the state parks system has qualified to host an AmeriCorps volunteer team. The federal program for men and women age 18-24 dispatches teams throughout the states for up to 11 months to perform about 1,700 hours of community service. The current nine-person team will spend more than six weeks in North Carolina’s state parks, participating in prescribed burns, preparing fire lines and removing exotic species and hardwoods in longleaf pine areas
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Meta
About me automatic widget
This is an automatic widget added on Third Bottom Widget box (Bottom Widget 3). To edit please go to Appearance > Widgets and choose 3rd widget from the top in area second called Bottom Widget 3. Title is also manageable from widgets as well.
Tag: fury
At first I was thinking of something vague like “wariness” or mundane like “dislike” or overwrought like “fear”. But no, “enmity” wins out. And it has the magical distinction of having shown up as a word of the day the day after I settled upon it.
When you’re growing up, you learn simple facts by which no explanations are attended. They Just Are. More than old wives tales (because they’re objectively demonstrable) but less than conclusions (because no one bothers to explain “why?”), such simple axioms have carried the day in our personal histories more than we care to even think about.
I’m sure you can think of a few, but consider examples such as
Approaching thunderstorms carry a heralding scent
“Autumn is in the air”
Adult cats avoid kittens
Adult dogs avoid puppies
But did you ever ask your parents or other adults——or even yourself—why those things are what they are? Probably you did ask others and got a “because”, or you asked yourself and were simply stumped or you pulled out the World-Book-Encyclopædia-Britannica and arrived at something close to—or at least close-enough-to—intellectual satisfaction.
Intellectual anything is near and dear to my heart—though less near than usual in these last fourteen months and therefore more dear in these last fourteen months—but as I get older and older (and isn’t 43 so much older and so much younger than you thought it was/would be when you were half that age?), the intellectualization of a thing becomes more and more an artifice, a synthetic supposition of the reality it attempts to approximate.
Death is an interesting one. Illness—especially one’s own—also interesting. Interesting because the distance between the idea of a thing and the actual thing is so much more vast than you ever, ever expected. By distance I most certainly do not mean incorrectness. Death is an ending and a shape to which the living must conform in order to continue on their own existences. Illness is much the same class of thing only varying in degree. Intellectualizations aren’t necessarily incorrect, they’re simply left wanting. And ironically, the better the match, intellectually to phenomenologically, the more poignant the distance: Being so right in one’s intellectual expectations and left so unbelievably wanting when the analogous reality occurs.
“When you’re dead, you’re dead.” That’s a James Lapine line from a Stephen Sondheim musical, if you can believe it. Unsung, aptly, and lacking bitterness, aptly, but in the book nonetheless.
We carry around with us our own sensibilities…
<br/>
<br/>
…and those sensibilities—my own sensibilities inform me—may account for so much more of our actions and reactions than any of us is willing to own up to.
This is not a bad thing. Sensibilities are, from a certain intellectual (yes, I know) perspective, learning that happens beneath, beside, astride, because-of, before, after, in-spite-of the bookish, garden-variety learning.
The adult cat senses the caprice of the kitten. The adult dog finds puppy-play unpredictable and moot. Cats and dogs may not play together, but they avoid together.
Humans avoid together, too. Or at least they should. When they can.
But human society mixes more, isolates less even as each of us feels more and more isolated. Instincts are devalued against intellect, positioned as competitors instead of symbionts. The results are skewed, sometimes to the point of catastrophe.
Maturity marches no longer in lock-step with age, but the instinctual expectation of such remains: we assume.
Assumptions are fragile and fall away silently. Consilience becomes archaic. We lose discernment, which is just another way of saying we regress: the kitten does not, cannot discern between play and attack. The puppy has no sense of boundaries.
The adult human is adult sometimes only in form and not function. The human child may lack a sophistication which imbues a wisdom his or her adult self is destined to lack.
These things begin at different times and progress/regress at different paces, often within the same individual. Fractionation leads to fractiousness and the road to hell is formed by the footfalls of such creatures.
There is no Hell, of course, except the one of Lore and Allegory, of Hyperbole and Admonition, but the imagery is of practical use.
People live with paradox because they must: survivors. Sometimes people live amidst paradox, willfully taking advantage of gray-space in a world increasingly unwilling to see things as anything but black or white: manipulators. Sometimes people resort to clever to spin up attractive paradoxes to their own advantage: monsters.
And some people are so damaged that they lose their way: the petulant.
This last group are the ones ultimately responsible for others having invented aphorisms like “no good deed goes unpunished” and “nice guys finish last”. Untruths, to be sure, unless the petulant are involved.
Petulance creates insult out of generosity, invents lies out of truth, twists candor into accusation and always, always lives the life of emotional vampire, the ultimate acquisitor.
There are many in our culture who turn terms of therapy into some a self-serving swamp of personal jingoism. These are the same people who weaponize grief and sorrow and generosity and good-intentions even when any of those things in their original good-will forms benefit themselves!
Now, I have been accused of many things: I’m too blunt, or too decorous. Too direct or too passive. Too generous or too stingily demanding. Too kind or too cruel. Too forgiving or too unrelenting. I have been accused of not being able to “just let it go”, whatever “it” was, and there is no opposite to that.
And like certain cultures whose counting mechanisms never evolved past “One. Two. Three. Many. Many. Many…”, such is my own openness to the world of “survivors”, “manipulators”, “monsters” and “the petulant”.
I am ever surprised when a gift results in thievery. Ever surprised when kindness is returned with abuse, forgiveness with with retribution.
Right now I am confounded by a smiling knife slipped with exacting precision between ribs in my back; I am also disturbed by the fact that the knife itself is only a minor part of the pain. No, filthy and base, it’s all down to matter of money and an ugly judgment based thereupon.
I was accused—and likely rightly so—of enabling bad behavior in another by covering his expenses. I accept this, refusing to question it because I am out of my depth in such matters. My accuser, however, is also out of his depth and did directly benefit financially from my “enabling”.
There is no wiggle room here; my guilt in enabling is clear and unqualified. Let that fact stand.
But I ask myself why I am furious at the accusation when I know it to be true. That my good intentions brought harm to another should—and does—bring guilt, not fury. That my good intentions brought benefit to my accuser? Well, that was part of my original impetus in covering expenses as the two people were financially interdependent! I was not helping just one person, but several, and was helping to keep an entire situation from fragmenting into a thousand sharp and ugly pieces.
So if the enabling-money was the problem in all of this, as the accuser pointed out, there’s a simple answer to reversing the damage: reverse the cashflow. This was not a suggestion to my accuser, it was a demand. Return the money that had caused the problem. Ungift the gift that had helped you out and had reduced your stress and had allowed your own life to proceed without significant disruption.
If the money was needed, as the situation more than implies, you’d expect his answer to be “I can’t” or “I don’t have the money to give you”. His actual answer? “It was your choice in giving the money and it’s not my responsibility to return it to you.”
Adult cats keep distance from the pin-pricks of kitten-claws and kitten-teeth. Adult dogs avoid the unpredictability of pups.
My fury? There’s no answer, there’s only the wisdom of cats and of dogs.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Thursday, 31 May 2012
3d Birds Wallpaper
3d Birds Wallpaper Biography
Perhaps the most striking thing about birds is that they come in all colors and in many sizes. Smaller birds are usually less expensive and their enclosures and play areas take up less space. Large birds need a lot more room and are more expensive to buy and maintain.
The sizes of birds varies greatly. Your pet can be a large bird like a macaw or cockatoo. It can be a medium large bird like the Amazon parrots, African Greys, or pigeons. You may want a moderate, medium sized bird like one of the conures, Ringneck Doves, or some of the bigger parakeets. Or you may want to start small, with a less expensive bird. Some of the most popular and readily available smaller birds are the budgerigars (called parakeets in the United States), cockatiels, lots of different types of finches, as well as canaries.
The many different colors birds come in is definitely intriguing as well. You may be drawn to a particular bird because of its attractive plumage. Fortunately color doesn't usually take any special consideration unless you get a bird that needs color feeding, like a red-factor canary, which needs special red foods to keep its coloring strong.
Bird Families:
An incredible thing about birds is that they are not all built the same. Because they are not built the same, different groups of birds or bird families have specialized traits that can be very desirable. Some people love talking parrots and others love the song of the canary. There are also people who love developing and showing different types of birds like canaries, pigeons, or even chickens. Other people enjoy working with birds of prey in the sport of falconry.
Bird Species:
You'll find that each of the different bird families, or groups of birds, will have a shared set of normal behaviors. This bird information is helpful in determining the traits that you want in your pet. However, each individual bird, though still having the characteristics of its group, also comes with its own unique characteristics and temperament.
Many parrots are known for their talking ability and some types are considered better talkers than others. But because each bird is unique, you could find your bird being a better talker than its group. Conversely you may choose a pet bird whose group is among the best of talkers, and your particular bird may only say a few phrases or not talk at all.
Bird Identification
The earliest known bird, Archaeopteryx lithographica, lived about 140 million years ago during the Jurassic period. It is known from fossil records discovered in slate quarries in southern Germany. These fossils show this early bird to be about the size of a long-tailed pigeon. Today there are more than 9000 bird species.
Depending on how birds are viewed or their use, bird identification can be approached is several different ways. Those studying them in a scientific manner identify birds by placing them in a taxonomical structure. While those that observe them or keep them as pets may identify birds in a more romantic or practical manner. Bird identification can actually include a number of different descriptions.
Scientific Classification
In the taxonomical classification system, all birds are members of the Aves Class. They are then divided into some 28-30 orders and each order has a scientific name to describe its group. Interestingly, more than half of all birds, about 60%, are contained within just a single order, the Order Passeriformes, sometimes referred to as the "perching birds".
Most of the scientific orders also have a functional name, or common name, associated with the group. For example, orders of aquatic birds are divided into such things as seabirds, diving birds, long-legged wading birds, and the more common waterfowl. Other orders contain flightless birds, shorebirds, ground dwelling birds, woodland birds, nocturnal birds, and a bunch more. As numerous and diverse as bird types are, the orders themselves can represent a single group, but more often contain a number of related types.
Ultimately in the taxonomic realm, each bird species itself has a single scientific name assigned to it. There is usually with one or more common names associated with it as well.
Romantic Names:
Besides the scientific names, common names, and functional naming of orders, there are also many romantic or fantastic names applied to groups of birds. You may hear birds described as lovebirds, exotic birds, endangered birds, tropical birds, blue birds, rainforest birds, wedding doves, and more.
Practical Names:
Perhaps the best know bird identification is with the use of practical names. Like the romantic names, these names also identify birds individually or as groups, but in a more pragmatic manner. Practical names are applied to those commonly kept as pets such as parrots, those seen when bird hunting or bird watching like the waterfowl, and also to birds kept for eggs or food such as poultry.
Pet Bird Identification:
Pet birds contain all sorts of bird types, and their names can include those from all the types above and more. Pet birds range from large to small parrots, like the Amazons, Macaws, Parakeets, Lovebirds, and more. Hard bills are birds like finches, canaries, pigeons, doves and other seed eaters. The soft bills are those that eat a fruit based diet, like lories, lorikeets, and toucans. There are a many other exotic birds kept as pets as well.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Q:
HTML/JS not parsing floats
I have an average and I need to update to the new average whenever the user changes his/her input. Unfortunately, HTML seems to only initiate a change if a int is put in, not a float. More specific:
<input type="number" step="0.01" ng-model="newRating" max="10" min="0" maxlength="2">
{{averageRating}}
AngularJS:
Parent Controller:
$scope.sumOfRatings is the previous sum of ratings,
newRating is the rating inputed,
$scope.volumeOfRatings is the previous number of rating:
$scope.$on('changeRating', function(eventName, newRating) {
$scope.averageRating = ($scope.sumOfRatings + newRating) / ($scope.volumeOfRatings + 1);
}
Code in Child Controller (parses the user input):
$scope.$watch('newRating', function(newRating, oldRating) {
$scope.$emit('changeRating', $scope.newRating);
});
However, this only works when numbers like 3, 10, 8, 5 are parsed. If I type in something like 0.33, or 9.8, this does nothing to the average. Sorry if I mist-typed any code, but I think the main problem is not the AngularJS code, but HTML or AngularJS recognizing changes. I did a console.log(newRating) in the $watch and found it never parsed any decimal inputs (like 8.55), only outputted 8.
How can I parse decimal changes in the HTML? I tried using an ng-repeat directive to no avail.
A:
Remove the max length from input tag.
<input type="number" step="0.01" ng-model="newRating" max="10" min="0">
{{averageRating}}
| null |
minipile
|
NaturalLanguage
|
mit
| null |
676 F.Supp.2d 1319 (2009)
Melissa J. SINNI, Plaintiff,
v.
SCOTTSDALE INSURANCE COMPANY, Defendant.
Case No. 6:08-cv-1787-Orl-31KRS.
United States District Court, M.D. Florida, Orlando Division.
December 18, 2009.
As Amended January 4, 2010.
*1321 Charles R. Steinberg, Charles R. Steinberg, PA, Cocoa, FL, for Plaintiff.
Julie Karen Linhart, Phelps Dunbar, LLP, Tampa, FL, for Defendant.
OPINION AND ORDER
GREGORY A. PRESNELL, District Judge.
This matter came before the Court without oral argument upon consideration of cross-motions for summary judgment filed by Plaintiff Melissa J. Sinni ("Plaintiff") and Defendant Scottsdale Insurance Company ("Scottsdale") and the parties' respective responses thereto (see Docs. 51, 53, 55 and 56).
I. Overview
This insurance dispute arises out of a slip-and-fall suit that culminated in a Coblentz agreement.[1] On May 3, 2007, Plaintiff sued Aaron P. Cullen and Absolutely Massage, Inc. (collectively, the "underlying defendants") in state court, alleging, inter alia, that she slipped and fell on a wet mulch walkway while leaving the underlying defendants' premises (see generally Doc. 2-2).[2] Count I of the Complaint asserted a claim for premises liability against Absolutely Massage, Inc. (Doc. 2-2, ¶¶ 20-25). Count II asserted that Absolutely Massage, Inc.'s general manager, Cullen, was negligent in maintaining Absolutely Massage, Inc.'s premises (Doc. 2-2, ¶¶ 10 and 26-32). Count III asserted a negligence claim against Absolutely Massage, Inc. predicated on respondeat superior and Cullen's negligence (Doc. 2-2, ¶¶ 33-35).
Approximately one year after filing suit, Plaintiff resolved all of her claims against the underlying defendants by entering into an "Assignment, Settlement Agreement and Covenant Not to Execute" (Doc. 17-43). Pursuant to the settlement agreement, Plaintiff and the underlying defendants agreed to the entry of a $300,000 judgment in favor of Plaintiff and against the underlying defendants (Doc. 17-44). They further agreed, however, that Plaintiff would not attempt to record or execute against the judgment (Doc. 17-44). Instead, *1322 the underlying defendants assigned all rights in their commercial general liability (CGL) policy to Plaintiff so that she could enforce the judgment against the underlying defendants' insurance carrier: Scottsdale (Doc. 17-44).
On May 21, 2008, the state court approved and entered the settlement agreement as a consent judgment (Doc. 18-2). Plaintiff thereafter filed an "amended"[3] complaint, seeking to enforce the consent judgment against Scottsdale (see Docs. 17-46 and 31-3). Scottsdale timely removed the case to this Court (Doc. 19).
The issue now before the Court is whether there was coverage for Plaintiff's claims in the state court action under the CGL policy that Scottsdale issued to the underlying defendants.[4] The Court has subject matter jurisdiction pursuant to 28 U.S.C. § 1332 (Doc. 19 at 3) and the parties agree that Florida substantive law is controlling.
II. Applicable Law
A. Summary Judgment
A party is entitled to summary judgment when it can show that there is no genuine issue as to any material fact. FED. R. CIV. P. 56(c); Beal v. Paramount Pictures Corp., 20 F.3d 454, 458 (11th Cir. 1994). Which facts are material depends on the substantive law applicable to the case. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). The moving party bears the burden of showing that no genuine issue of material fact exists. Celotex Corp. v. Catrett, 477 U.S. 317, 323, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986); Clark v. Coats & Clark, Inc., 929 F.2d 604, 608 (11th Cir.1991); Watson v. Adecco Employment Servs., Inc., 252 F.Supp.2d 1347, 1351-52 (M.D.Fla.2003). In determining whether the moving party has satisfied its burden, the court considers all inferences drawn from the underlying facts in a light most favorable to the party opposing the motion, and resolves all reasonable doubts against the moving party. Anderson, 477 U.S. at 255, 106 S.Ct. 2505.
When a party moving for summary judgment points out an absence of evidence on a dispositive issue for which the non-moving party bears the burden of proof at trial, the non-moving party must "go beyond the pleadings and by [its] own affidavits, or by the depositions, answers to interrogatories, and admissions on file, designate specific facts showing that there is a genuine issue for trial." Celotex Corp., 477 U.S. at 324-25, 106 S.Ct. 2548 (internal quotations and citations omitted). Thereafter, summary judgment is mandated against the non-moving party who fails to make a showing sufficient to establish a genuine issue of fact for trial. Id. at 322, 324-25, 106 S.Ct. 2548; Watson, 252 F.Supp.2d at 1352. The party opposing a motion for summary judgment must rely on more than conclusory statements or allegations unsupported by facts. Evers v. Gen. Motors Corp., 770 F.2d 984, 986 (11th Cir.1985) ("conclusory allegations without *1323 specific supporting facts have no probative value") (citations omitted); Broadway v. City of Montgomery, Ala., 530 F.2d 657, 660 (5th Cir.1976).
With respect to determining insurance coverage issues, summary judgment is generally appropriate inasmuch as the construction and legal effect of a written contract are matters of law to be determined by the Court. See, e.g., Northland Cas. Co. v. HBE Corp., 160 F.Supp.2d 1348, 1358 (M.D.Fla.2001) ("Summary judgment is appropriate in declaratory judgment actions seeking a declaration of coverage when the insurer's duty, if any, rests solely upon the applicability of the insurance policy, the construction and effect of which is a matter of law.") (citations omitted).
B. Insurance
In Florida, an insurer's duty to defend is distinct from, and broader than, the duty to indemnify. Lime Tree Vill. Cmty. Club Ass'n, Inc. v. State Farm Gen. Ins. Co., 980 F.2d 1402, 1405 (11th Cir. 1993) (citing Baron Oil Co. v. Nationwide Mut. Fire Ins. Co., 470 So.2d 810, 813-14 (Fla. 1st DCA 1985)). The duty to defend is determined by examining the allegations in the underlying complaint against the insured. Id. (citing Nat'l Union Fire Ins. Co. v. Lenox Liquors, Inc., 358 So.2d 533, 536 (Fla.1977)); see also State Farm Fire & Cas. Co. v. CTC Dev. Corp., 720 So.2d 1072, 1077 n. 3 (Fla.1998). The insurer must defend when the complaint alleges facts that fairly and potentially bring the suit within coverage. Jones v. Fla. Ins. Guar. Ass'n, Inc., 908 So.2d 435, 442-43 (Fla.2005). The allegations in the complaint control even when the actual facts are inconsistent with the allegations in the complaint. Id. at 443. Furthermore, once a court finds that there is a duty to defend, the duty will continue "even though it is ultimately determined that the alleged cause of action is groundless and no liability is found within the policy provisions defining coverage." Baron Oil Co., 470 So.2d at 814. In exceptionally rare cases, however, where the complaint omits a reference to an uncontroverted fact that, if pled, would have clearly placed the claim outside the scope of coverage, equity may relieve an insurer from its duty to defend. See Underwriters at Lloyds London v. STD Enters., Inc., 395 F.Supp.2d 1142 (M.D.Fla.2005) [hereinafter "Underwriters"]; Nationwide Mut. Fire Ins. Co. v. Keen, 658 So.2d 1101 (Fla. 4th DCA 1995) [hereinafter "Keen"]; but see First Specialty Ins. Corp. v. 633 Partners, Ltd., 300 Fed.Appx. 777, 786 (11th Cir.2008) (suggesting that, unless it is "manifestly obvious" to the parties that the actual facts put the plaintiff's claims outside the scope of coverage, Underwriters and Keen should be regarded as aberrations) (citations omitted) (unpublished).[5]
An insurer's duty to indemnify is narrower than its duty to defend and must be determined by analyzing the policy coverages in light of the actual facts in the underlying case. CTC Dev. Corp., 720 So.2d at 1077 n. 3 (citing Hagen v. Aetna Cas. & Sur. Co., 675 So.2d 963, 965 (Fla. 5th DCA 1996)); see also, e.g., Auto Owners Ins. Co. v. Travelers Cas. & Sur. Co., 227 F.Supp.2d 1248, 1258 (M.D.Fla.2002). In construing insurance policies, courts should read the policy as a whole, endeavoring to give every provision its full meaning and operative effect. U.S. Fire Ins. Co. v. J.S.U.B., Inc., 979 So.2d 871, 877 (Fla.2007) (citation). Exclusionary clauses are typically read strictly and in a manner that affords the insured the broadest possible coverage. Indian Harbor Ins. Co. v. Williams, 998 So.2d 677, 678 (Fla. 4th DCA 2009).
*1324 C. Coblentz Agreements
In Florida, a party seeking to recover under a Coblentz agreement must prove: (1) coverage; (2) a wrongful refusal to defend; and (3) that the settlement was objectively reasonable and made in good faith. Chomat v. N. Ins. Co. of N.Y., 919 So.2d 535, 537 (Fla. 3d DCA 2006) (quoting Quintana v. Barad, 528 So.2d 1300, 1301 n. 1 (Fla. 3d DCA 1988) (internal citation omitted)); see also, e.g., Shook v. Allstate Ins. Co., 498 So.2d 498 (Fla. 4th DCA 1986). The party seeking to recover has the initial burden of proving that its underlying claim against the insured was within the coverage of the policy. See, e.g., Steil v. Fla. Physicians' Ins. Reciprocal, 448 So.2d 589, 592 (Fla. 2d DCA 1984); Keller Indus., Inc. v. Employers Mut. Liab. Ins. Co. of Wis., 429 So.2d 779, 780 (Fla. 3d DCA 1983).[6]
The determination of coverage is a condition precedent to any recovery against an insurer pursuant to a Coblentz agreement. Spencer v. Assurance Co. of Am., 39 F.3d 1146, 1149 (11th Cir.1994); Farmer v. Liberty Mut. Fire Ins. Co., Case No. 06-61477-CIV, 2007 WL 2028842 at *3 (S.D.Fla. July 9, 2007); Steil, 448 So.2d at 592. As the Eleventh Circuit has observed:
Florida law clearly states that liability of an insurer depends upon whether the insured's claim is within the coverage of the policy. This remains true even when the insurer has unjustifiably failed to defend its insured in the underlying action. A determination of coverage, therefore, is a condition precedent to any recovery against an insurer [under a Coblentz agreement].
Spencer, 39 F.3d at 1149 (11th Cir.1994) (citing Steil, 448 So.2d at 592; Fla. Farm Bureau Mut. Ins. Co. v. Rice, 393 So.2d 552 (Fla. 1st DCA 1980); and Central Nat'l Ins. Co. v. Gonzalez, 295 So.2d 694, 696 (Fla. 3d DCA 1974)). Indeed, the mere entry of a consent judgment does not establish coverage and an insurer's unjustifiable failure to defend the underlying action does not estop the insurer from raising coverage issues in a subsequent suit to satisfy a consent judgment entered pursuant to a Coblentz agreement. Spencer, 39 F.3d at 1149.
Accordingly, there are two prongs to the coverage element in an action to recover under a Coblentz agreement: (1) the facts alleged in the underlying complaint must state a claim that fell within the coverage of the policy (i.e., that the insurer had a duty to defend); and (2) notwithstanding the allegations in the underlying complaint or stipulated facts in the consent judgment, the plaintiff's underlying claims must actually come within the coverage of the policy (i.e., on the merits, the insurer has a contractual duty to indemnify). See, e.g., Spencer v. Assurance Co. of Am., Case No. 91-50255-RV, 1993 WL 761408 (N.D.Fla. June 8, 1993) (concluding that, although insurer had a duty to defend based on the allegations in the underlying complaint, plaintiffs could not recover under Coblentz agreement because the actual facts were such that plaintiffs' claims did not come within the coverage of the policy), aff'd, 39 F.3d 1146 (11th Cir.1994); Farmer, Case No. 06-61477-CIV, Order at Doc. 78-2 (S.D.Fla. July 18, 2007) (vacating 2007 WL 2028842 (S.D.Fla. July 9, 2007) and granting insurer's motion for *1325 summary judgment because allegations in underlying complaint failed to establish that insurer had a duty to defend and facts stipulated to in Coblentz agreement could not be used to establish coverage). With respect to the second prong, this Court has previously observed:
While the duty to defend is broad and must be determined solely by the allegations in the complaint against the insured, the duty to indemnify is more narrow and must be measured by the facts adduced at trial or developed through discovery ... Therefore, notwithstanding the allegations against the insured, if the facts show that the insured's liability stems from a claim for which no coverage is provided under the policy, the insurer owes no duty of indemnification. This premise holds true even where, as here, the insured's liability was arrived at through a settlement of the action against the insured, because a settlement does not, by itself, obligate the insurer to pay for a non-covered claim. Instead, the insurer's duty to indemnify a settlement obligation must be measured by the facts... extant at the time the settlement was reached.
Travelers Indem. Co. of Ill. v. Royal Oak Enters., Inc., 344 F.Supp.2d 1358, 1366 (M.D.Fla.2004) (Hodges, J.)
III. Undisputed Facts
A. Background
Plaintiff began working for Absolutely Massage, Inc. as a massage therapist in September of 2005 (Doc. 51-32 at 5, Sinni Dep. at 15). On November 14, 2005, Plaintiff clocked out and left Absolutely Massage, Inc.'s building (Doc. 51-32 at 6, Sinni Dep. at 18). However, as Plaintiff attempted to proceed through a gate that led to an adjacent parking lot in which she had parked her car,[7] she slipped and fell on a pile of wet mulch located on her employer's premises (Doc. 51-32 at 6, Sinni Dep. at 18).[8]
As a result of her slip-and-fall injuries,[9] Plaintiff filed a workers' compensation claim[10] and, as early as December 2, 2005, *1326 began receiving payments from Absolutely Massage, Inc.'s workers' compensation carrier, The Hartford (Doc. 51-32 at 9, Sinni Dep. at 32; Doc. 53-16 at 8). Between December 2005 and April 2007 i.e., prior to filing the underlying tort action Plaintiff received nearly $18,500 in payments from The Hartford (Doc. 53-16 at 1-9). In addition to the payments that were made directly to Plaintiff, The Hartford also paid more than $10,000 to Plaintiff's medical providers (Doc. 53-16 at 10-23). Furthermore, in October 2008, Plaintiff settled her workers' compensation claim with The Hartford for a gross lump-sum payment of $20,000 (see Doc. 51-43).[11] As of October 2008, all of Plaintiff's medical expenses were paid for by the workers' compensation benefits that Plaintiff received in the settlement with The Hartford (Doc. 51-32 at 9 and 14, Sinni Dep. at 30 and 53).
On June 16, 2006, approximately seven months after Plaintiff had first started receiving workers' compensation benefits from The Hartford, Plaintiff's counsel sent a letter and "ACORD[12] General Liability Notice of Occurrence/Claim" directly to Scottsdale.[13] Scottsdale construed the a letter and "Notice of Occurrence" as a claim, sent a reservation of rights letter to its insureds on June 29, 2006, and assigned an independent adjuster to investigate (Doc. 53-20 at 4-5, Stutheit Dep. at 11-13; Doc. 53-25). Scottsdale's investigation revealed that Plaintiff was Absolutely Massage, Inc.'s employee, that she was acting within the scope and course of her employment at the time of her slip-and-fall, and that she had been receiving workers' compensation benefits (Doc. 53-26). On March 5, 2007, Scottsdale sent a declination letter to Absolutely Massage, Inc. in which it specifically disclaimed coverage and refused to "participate in the defense or indemnification" of Plaintiff's claim due to, inter alia, the workers' compensation and employer's liability exclusions in its policy (Doc. 53-26 at 10). Two months later, Plaintiff filed *1327 her tort suit against Cullen and Absolutely Massage, Inc.
Although Plaintiff filed suit on May 3, 2007, according to Scottsdale, it did not learn of the litigation until February 8, 2008,[14] when counsel for Absolutely Massage, Inc., Mr. Alfred Truesdell, sent Scottsdale a letter demanding that it defend Absolutely Massage, Inc. and informing Scottsdale that his client was contemplating entering into a Coblentz agreement (see generally Doc. 53-27). On February 27, 2008, Mr. Truesdell forwarded a copy of the complaint to Scottsdale (Doc. 53-27 at 5). On March 25, 2008, Scottsdale sent a letter to Mr. Truesdell in which it acknowledged receipt of the complaint and stated: "[W]e are reviewing the complaint... [and] will provide you with our position with respect to coverage in the near future" (Doc. 53-29 at 1). On April 9, 2008, Scottsdale sent another letter to Mr. Truesdell, stating: "We have referred this matter to coverage counsel and hope to have their opinion within the next two weeks. I will be in contact with you then to advise whether we will agree to provide a defense to Absolutely Massage" (Doc. 53-30). On April 18, 2008, Mr. Truesdell sent another letter to Scottsdale, stating that he also represented Cullen and that Cullen demanded a defense (Doc. 53-27 at 12-13). His letter goes on to warn: "If we do not receive written confirmation that a defense will be provided to my clients from your offices by 5:00 p.m. on April 23, 2008, we will construe that to mean that Scottsdale ... has decided to forgo providing a defense in this matter" (Doc. 53-27 at 12-13). When Scottsdale failed to comply with Mr. Truesdell's deadline, the underlying defendants executed the Coblentz agreement on May 15th, 2008 (Doc. 17-44 at 8). The state court reduced the Coblentz agreement to a consent judgment on May 21, 2008. On or about May 29, 2008, one of Scottsdale's adjusters called Mr. Truesdell and told him that Scottsdale would undertake the defense of his clients pursuant to a reservation of rights (Doc. 53-28 at 35).[15]
At no time between the filing of Plaintiff's suit against the underlying defendants and the entry of the consent judgment did Scottsdale ever explicitly refuse to provide a defense or deny coverage. Nor, however, did it ever provide a defense.
B. Relevant Policy Provisions
Although the Scottsdale CGL policy contains three types of coverage, Plaintiff discusses her claims only within the context of Coverage A (Doc. 51 at 6-8). In pertinent part, Coverage A provides:
a. We will pay those sums that the insured becomes legally obligated to pay as damages because of "bodily injury" or "property damage" to which this insurance applies. We will have the right and duty to defend the insured against any "suit" seeking those damages. However, we will have no duty to defend *1328 the insured against any "suit" seeking damages for "bodily injury" or "property damage" to which this insurance does not apply. We may, at our discretion, investigate any "occurrence" and settle any claim or "suit" that may result ...
b. This insurance applies to "bodily injury" and "property damage" only if:
(1) The "bodily injury" or "property damage" is caused by an "occurrence"... [and]
(2) The "bodily injury" or "property damage" occurs during the policy period....
"Bodily injury" means bodily injury, sickness or disease sustained by a person, including death resulting from any of these at any time....
"Occurrence" means an accident, including continuous or repeated exposure to substantially the same general harmful conditions....
"Suit" means a civil proceeding in which damages because of "bodily injury" ... to which this insurance applies are alleged....
(Doc. 14-2 at 8, 19 and 22).
The following exclusions are also contained in the Scottsdale policy:
This insurance does not apply to:
d. Workers' Compensation and Similar Laws
Any obligation of the insured under a workers' compensation, disability benefits or unemployment compensation law or any similar law.
e. Employer's Liability
"Bodily injury" to:
(1) An "employee" of the insured arising out of and in the course of:
(a) Employment by the insured; or
(b) Performing duties related to the conduct of the insured's business ...
This exclusion applies:
(1) Whether the insured may be liable as an employer or in any other capacity; and
(2) To any obligation to share damages with or repay someone else who must pay damages because of the injury....
(Doc. 14-2 at 9).
Finally, the Scottsdale policy provides:
Duties In the Event of Occurrence, Offense, Claim or Suit
a. You must see to it that we are notified as soon as practicable of an "occurrence" or an offense which may result in a claim. To the extent possible, notice should include:
(1) How, when and where the "occurrence" or offense took place;
(2) The names and addresses of any injured persons and witnesses; and
(3) The nature and location of any injury or damage arising out of the "occurrence" or offense.
b. If a claim is made or "suit" is brought against any insured, you must:
(1) Immediately record the specifics of the claim or "suit" and the date received; and
(2) Notify us as soon as practicable.
You must see to it that we receive written notice of the claim or "suit" as soon as practicable....
(Doc. 14-2 at 17).
IV. Discussion
A. Duty to Defend
Scottsdale appears to concede that the allegations in Plaintiff's underlying complaint potentially brought the suit within coverage (see Doc. 53 at 14-15). Under the exceptional circumstances of this case, however, Scottsdale contends that it should be relieved of its duty to defend. Specifically, Scottsdale points to *1329 the deliberate omission of any reference in the underlying complaint to Plaintiff's uncontroverted receipt of workers' compensation benefits and employment status. Furthermore, Scottsdale argues that it was prejudiced by its insureds' breach of their duty to promptly forward a copy of the underlying complaint to Scottsdale until nine months after they were served and just two and half-months before the insureds entered into the Coblentz agreement (Doc. 53 at 17 and 20).
The Court agrees that the circumstances of this case are exceptional and that Scottsdale should be relieved of its duty to defend. As discussed in more detail, infra, it was manifestly obvious that the actual facts put Plaintiff's claims outside the scope of coverage. When Plaintiff filed suit, she had already applied for and was receiving workers' compensation benefits for the very slip-and-fall injuries that comprised the only basis for her underlying action. Had Plaintiff not concealed that uncontroverted fact from her complaint, her claims would clearly have been barred by workers' compensation immunity and outside the scope of coverage. Instead, after Scottsdale had received Plaintiff's "Notice of Occurrence" and denied coverage, Plaintiff decided to recast her claims and mislead the court by alleging that she was simply a "business invitee." The insureds then kept Scottsdale in the dark for more than nine months after they were served.[16] When they finally gave Scottsdale notice of the suit, they immediately indicated that they were contemplating a Coblentz agreement.[17] Less than three months later, while Scottsdale was still attempting to determine whether it had a duty to defend, the insureds executed the Coblentz agreement.
In addition, then, to the omission of an uncontroverted fact in the complaint that would have justified Scottsdale's denial of a defense, Scottsdale's right to assess Plaintiff's claims and exercise control over the litigation before the parties entered into the Coblentz agreement was significantly impaired. Based on these exceptional circumstances, the Court concludes that Scottsdale was relieved of its duty to defend and is therefore entitled to a judgment as a matter of law that it has no obligation to reimburse Plaintiff or its insureds for any defense costs or attorneys' fees incurred in the underlying litigation.
B. Duty to Indemnify
Even assuming, arguendo, that Scottsdale had a duty to defend, the Court concludes that Plaintiff's claims did not come within the coverage of Scottsdale's CGL policy. As the Third District Court has recognized, the standard workers' compensation and employer's liability exclusions in CGL policies such as Scottsdale's preclude coverage for Plaintiff's injuries. Fla. Ins. Guar. Ass'n, Inc. v. Revoredo, 698 So.2d 890, 892 (Fla. 3d DCA 1997) ("[T]he only coverage intended [in a CGL policy], and for which the premium has been paid, is the liability of the insured to the public, as distinguished from liability to the insured's employees whether or not they are protected by the workers' compensation law").
As a threshold matter, it is well-established that workers' compensation is the sole and exclusive remedy available to a worker injured in a manner that falls within the broad scope and polices of Florida's workers' compensation statute. Byrd v. Richardson-Greenshields Sec., Inc., 552 *1330 So.2d 1099, 1100 (Fla.1989); FLA. STAT. § 440.11 (an employer's liability to an employee "shall be exclusive and in place of all other liability, including vicarious liability...."); see also, e.g., Eller v. Shova, 630 So.2d 537, 539 (Fla.1993); Seaboard Coast Line R.R. Co. v. Smith, 359 So.2d 427, 429 (Fla.1978). Under the statute, an employer is obligated to compensate an employee for injuries "arising out of work performed in the course and the scope of employment." FLA. STAT. § 440.09(1). In keeping with the liberal construction placed on this statute, Byrd, 552 So.2d at 1100, however, courts have broadly construed "arising out of" and "in the course and the scope of employment" to include claims that have only a sufficient nexus to the workplace. See, e.g., Perez v. Publix Supermarkets, Inc., 673 So.2d 938 (Fla.3d DCA 1996) (holding that employee who slipped and fell after clocking out and while leaving employer's business suffered injury "arising out of work performed in the course and the scope of employment" and that employee's negligence claim was barred because workers' compensation was her exclusive remedy); Vigliotti v. K-mart Corp., 680 So.2d 466 (Fla. 1st DCA 1996) (holding that employee who clocked out and was exiting employer's premises when she slipped and fell suffered injury "in the course and the scope of employment"); Jenkins v. Wilson, 397 So.2d 773 (Fla. 1st DCA 1981) (finding that employee who had clocked out and was accosted and raped while crossing alley leading to parking lot suffered an injury "arising out of" and "in the scope of" her employment).
It is undisputed that Plaintiff received significant workers' compensation benefits for the very injuries for which she subsequently sought relief in the underlying tort action. Indeed, Absolutely Massage, Inc.'s workers' compensation carrier began making payments directly to Plaintiff and her medical providers as early as December 2, 2005 and later settled Plaintiff's entire claim for a lump-sum payment. Notwithstanding these payments, Plaintiff then decided to sue her employer and fellow employee in tort for the very same injuries. While Plaintiff concealed her employment status and workers' compensation claim, it is evident that Plaintiff's injuries arose out of and in the course of her employment and that her underlying claims amounted to a workers' compensation obligation within the meaning of Scottsdale's CGL policy. See, e.g., Naranja Rock Co. v. Dawal Farms, 74 So.2d 282, 286 (Fla. 1954); Kramer v. Palm Beach County, 978 So.2d 836 (Fla. 1st DCA 2008); Perez, 673 So.2d at 939-940 (holding that workers' compensation was employee's exclusive remedy where slip-and-fall injuries were suffered after clocking out and while leaving employer's premises); see also Maas Bros. v. Peo, 498 So.2d 657 (Fla. 1st DCA 1986).
Plaintiff, however, contends that the foregoing is of no moment (see, e.g., Doc. 55 at 5). Because Scottsdale failed to defend its insureds, Plaintiff argues that Scottsdale waived all of its affirmative defenses, including, in particular, the defense of workers' compensation immunity. Furthermore, Plaintiff argues that the workers' compensation exclusion in Scottsdale's CGL policy a coverage defense that is separate and apart from the affirmative defense of workers' compensation immunity is inapposite because the consent judgment is not "an obligation under a workers' compensation law" (Doc. 55 at 5). Finally, Plaintiff argues that the employer's liability exclusion also does not apply. In support of these contentions, Plaintiff relies on Wright v. Hartford Underwriters Ins. Co., 823 So.2d 241 (Fla. 4th DCA 2002) [hereinafter Wright].
In Wright, the plaintiff suffered a workplace injury and filed a workers' compensation claim. 823 So.2d at 241. After *1331 settling his workers' compensation claim with his employer's insurance carrier, the plaintiff then brought a tort action alleging that his fellow employee was grossly negligent and that his employer was vicariously liable for the fellow employee's negligence. Id. After defendants' CGL insurer (which was the same carrier that had settled the plaintiff's workers' compensation claim) refused to defend the tort action, the plaintiff and defendants entered into a Coblentz agreement. Id. at 241-42. The Coblentz agreement was then reduced to a consent judgment and the plaintiff sued the CGL insurer. Id. at 242. The trial court granted summary judgment to the insurer, ruling that the defendants in the underlying action would have been entitled to workers' compensation immunity. Id. In reversing, the Fourth District Court held:
Workers' compensation immunity is a defense that [the defendants] could have raised in the [underlying action]. Because [the insurer] refused to defend its insured, it is bound by the settlement waiving the defense of workers' compensation immunity and may not assert that defense against [the plaintiff's] claim for policy benefits to satisfy the [consent judgment]. The insurer is now bound by the settlement and may not re-litigate the issue of liability by raising any affirmative defense that could have been raised in the [underlying action].
Id. at 242-43 (citing Ahern v. Odyssey Re (London) Ltd., 788 So.2d 369, 371 (Fla. 4th DCA 2001) and Indep. Fire Ins. Co. v. Paulekas, 633 So.2d 1111, 1114 (Fla. 3d DCA 1994)) (footnotes omitted). In dicta, however, the Fourth District Court went on to note:
Our disposition of the workers [sic] compensation immunity defense addresses the only basis for the trial court's grant of summary judgment. On remand the trial court will be confronted with the remaining coverage issue-e.g., whether its employer's liability policy coverage... extended to include [the plaintiff's underlying action]. We do point out, however, that the workers [sic] compensation exclusion in the [insurance policy], relied upon by the trial court below, does not apply to [the plaintiff's] civil action because the settlement judgment was not an "obligation imposed by worker's [sic] compensation" law. Rather, the judgment arose from the claims in the civil action and the settlement agreement... neither of which involve obligations imposed by workers [sic] compensation law.
Id. at 243.
On its face, the actual holding in Wright is that an insurer cannot avoid its duty to indemnify by raising an affirmative defense that its insured failed to raise or could have raised in the underlying action. Id. at 242; see also Gallagher v. Dupont, 918 So.2d 342 (Fla. 5th DCA 2005) (holding, inter alia, that insurer which refused to defend its insured in underlying action could not raise an affirmative defense to insured's liability which could have been raised in underlying action); Ahern, 788 So.2d 369 (holding that insurer in a Coblentz action could not contest the liability of its insured in prior underlying action). Wright's holding has no bearing, however, on an insurer's ability to raise coverage defenses or exclusions that appear on the face of the policy. On the contrary, an insurer faced with a Coblentz action has the right to litigate its contractual duty to indemnify on the actual facts of the underlying litigation just as it would in an action against its insured. See, e.g., Spencer, 39 F.3d at 1149 (citing, inter alia, Steil, 448 So.2d at 592).
When considering the effect of a Coblentz agreement where the insurer has breached its duty to defend, it is important to recognize the different relationships involved. On the one hand is the relationship *1332 in tort between the underlying plaintiff and the insured defendant. On the other hand, the insured's claim for indemnification of the tort obligation is governed by contract.
The insurance policy provides for coverage (indemnification) of amounts that the insured "becomes legally obligated to pay as damages because of bodily injury..." (Doc. 14-2 at 8). When the insured (or its assignee) sues for indemnification, the insurer cannot obviate its contractual liability by raising affirmative defenses (e.g., the statute of limitations, comparative fault, etc.) that might have been available to its insured in the tort case. Having wrongfully failed to defend its insured, the insurer has waived any such defenses and the underlying tort obligation has been established.
Conversely, when the insured or Coblentz plaintiff sues the insurer in contract for indemnification, policy exclusions may be raised as defenses, because those issues have neither been litigated nor waived. And this is true even if these policy defenses mirror defenses (e.g., workers' compensation immunity) that could have been raised to defend the tort claim itself. The labels attached to the defenses are irrelevant. One defense is contractual; the other sounds in tort. Similarly, coverage is determined by looking at the policy and the actual factsnot the labels attached to the underlying tort obligation. Because the parties' relationship is strictly contractual at the indemnification stage, neither the insured nor a Coblentz plaintiff can re-write the policy to provide benefits for non-covered injuries that have simply been cloaked in causes of action for which there superficially appears to be coverage.
To the extent that dicta in Wright suggests otherwise, that dicta fails to recognize this basic distinction in the relationships between the parties. The Court is therefore firmly convinced that the Florida Supreme Court would not follow Wright for the proposition cited to therein by Plaintiff.[18]
Even assuming, however, that the workers' compensation exclusion does not apply, the employer's liability exclusion in Scottsdale's CGL policy precludes coverage for Plaintiff's claims. According to the First District Court, the purpose of workers' compensation exclusions is to exclude coverage of "those employees protected by the workers' compensation law, whereas the language of [the employer's liability exclusion] ... exclude[s] liability for injury to employees generally. Although the two may overlap to a certain degree, they are not to be read together to exclude only those employees protected by workers' compensation." Greathead v. Asplundh Tree Expert Co., 473 So.2d 1380, 1383 (Fla. 1st DCA 1985) (second emphasis added) (citation omitted); see also Revoredo, *1333 698 So.2d 890. Furthermore, the Florida Supreme Court has held that, in the context of CGL policies, courts should read "arising out of" very broadly. Taurus Holdings, Inc. v. U.S. Fid. & Guar. Co., 913 So.2d 528, 535-40 (Fla.2005) (citations omitted). Thus, an employer's liability exclusion for bodily injury to an employee "arising out of and in the course of employment" encompasses claims that are potentially broader than workers' compensation obligations and may have only a limited causal relationship to employment. See, e.g., XL Ins. Am., Inc. v. Ortiz, 673 F.Supp.2d 1331, Case No. 09-20630-CV, 2009 WL 3739072 at *10 (S.D.Fla. Nov. 6, 2009) (applying identical employer's liability exclusion in CGL policy and holding that injuries resulting from the misuse of company time, such as "horseplay" or other "tomfoolery" that may not have given rise to an obligation under Florida's workers' compensation statute, were nevertheless outside the scope of coverage).
Here, there can be no question that Plaintiff was on Absolutely Massage, Inc.'s premises not as "business invitee" but as an employee. She had just finished her workday and was on her way to the parking lot when she slipped and fell. But-for her employment and her employer's directive to use a particular egress and parking lot, her injuries would not have occurred. Irrespective, then, of whether her injuries amounted to an obligation under Florida's workers' compensation statute, the employer's liability exclusion in Scottsdale's CGL policy also precludes coverage.
Accordingly, the Court concludes that Plaintiff's injuries comprised "an obligation of the insured under a workers' compensation law" and "arose out of and in the course of her employment" by Scottsdale's insured. Therefore, the workers' compensation and employer's liability exclusions in Scottsdale's CGL policy preclude coverage and Scottsdale is entitled to a judgment as a matter of law that it has no duty to indemnify Plaintiff or its insureds for Plaintiff's injuries or for the consent judgment entered in the underlying tort action.
V. Conclusion
For the foregoing reasons, it is ORDERED that Plaintiff Melissa J. Sinni's Motion for Summary Judgment (Doc. 51) is DENIED and Defendant Scottsdale Insurance Company's Motion for Summary Judgment (Doc. 53) is GRANTED.
It is further ORDERED and ADJUDGED that the Clerk of the Court shall enter judgment in favor of Defendant, Scottsdale Insurance Company, and against Plaintiff, Melissa J. Sinni, declaring that:
1. Defendant had no duty to defend Aaron P. Cullen or Absolutely, Massage, Inc. in the state court action styled Melissa J. Sinni v. Aaron P. Cullen and Absolutely Massage, Inc., Case No. 2007-CA-005165-O, Ninth Judicial Circuit, in and for Orange County, Florida, and has no duty to reimburse any party or assignee for defenses costs and attorneys' fees incurred therein; and
2. Defendant has no duty to indemnify Plaintiff, Aaron P. Cullen or Absolutely, Massage, Inc. for Plaintiff's slip-and-fall injuries of November 14, 2005 or for the consent judgment entered in the state court action styled Melissa J. Sinni v. Aaron P. Cullen and Absolutely Massage, Inc., Case No. 2007-CA-005165-O, Ninth Judicial Circuit, in and for Orange County, Florida.
The Clerk of the Court is further directed to close the file.
NOTES
[1] In certain circumstances, where an insurer has wrongfully refused to defend its insured and there is coverage under the policy, the insurer may be bound by the terms of a negotiated final consent judgment entered against its insured. See generally Coblentz v. Am. Sur. Co. of N.Y., 416 F.2d 1059 (5th Cir.1969) and its progeny.
[2] Although omitted from her Complaint, as discussed further, infra, it is undisputed that Plaintiff's employer was Absolutely Massage, Inc.; that Plaintiff's slip-and-fall occurred on her employer's premises while she was attempting to walk to her car after leaving work; and that Plaintiff received workers' compensation benefits for her slip-and-fall injuries.
[3] Although the new complaint was styled as an "Amended Complaint" and was ostensibly brought against Aaron P. Cullen, Absolutely Massage, Inc. and Scottsdale, it does not contain any of the former counts asserted against the underlying defendants (or for that matter, any claims whatsoever against the underlying defendants) (Doc. 17-46). Instead, Plaintiff simply alleged that she was "entitled to bring this action against SCOTTSDALE" to collect on the consent judgment entered by the state court (Doc. 17-46, ¶ 10). Consequently, the Court dismissed Cullen and Absolutely Massage, Inc. (Doc. 42).
[4] On August 17, 2009, the Court granted the parties' Agreed Motion to Bifurcate and directed the parties to file dispositive motions on the issue of coverage by no later than October 30, 2009 (Doc. 50). Those motions became ripe on November 2, 2009.
[5] Pursuant to 11th Cir. R. 36-2, "Unpublished opinions are not considered binding precedent, but they may be cited as persuasive authority."
[6] Once coverage is established, the claimant must then "assume the burden of initially going forward with the production of evidence sufficient to make a prima facie showing of reasonableness and lack of bad faith, even though the ultimate burden of proof will rest upon the carrier." Steil, 448 So.2d at 592. As noted, supra, the issues of reasonableness and good faith (or lack of bad faith) are not currently before the Court.
[7] Absolutely Massage, Inc. instructed Plaintiff and other employees to park their cars in the parking lot adjacent to the gate at the back of its premises (Doc. 51-32 at 6, Sinni Dep. at 19); (Doc. 53-13 at 17, Cullen Dep. at 16-17). Furthermore, employees would regularly exit through the back of the premises and traverse the path that led through the gate where Plaintiff slipped and fell (Doc. 53-13 at 17, Cullen Dep. at 17).
[8] It is undisputed that Plaintiff's fall occurred on Absolutely Massage, Inc.'s property (Doc. 51-32 at 6, Sinni Dep. at 19). In addition to Plaintiff's deposition testimony, the tort claims upon which the consent judgment was entered were each predicated on the fact that Plaintiff fell on Absolutely Massage, Inc.'s property (Doc. 2, ¶¶ 17-18).
[9] Plaintiff testified that she was diagnosed with one or more herniated and bulging discs in her neck, and tendinitis in one of her wrists, as a result of her slip-and-fall (Doc. 51-32 at 7, Sinni Dep. at 24-25). Plaintiff also testified that, prior to her slip-and-fall, she suffered a herniated disc in her back as a result of an automobile accident for which she received $100,000 in settlement proceeds (Doc. 51-32 at 7, Sinni Dep. at 23).
[10] On January 6, 2006, Plaintiff filed her petition for workers' compensation benefits with the State of Florida's Division of Administrative Hearings, Office of the Judges of Compensation Claims. See Petition for Benefits filed on 1/6/2007, available at http://www.jcc. state.fl.us/JCC/searchJCC/, Case No. 06-000191 WJC (Condry, J.) [hereinafter the "WC Petition" or "Workers' Compensation Petition"]. Plaintiff was represented by counsel in the workers' compensation proceeding (WC Petition at 1). On the portion of the form directing the claimant to provide a "Description of Work Being Performed When Injury Occurred," Plaintiff certified that she "Performs [sic] duties as a massage therapist" (WC Petition at 1). As to the "Description of Accident and Parts of Body Injured," Plaintiff certified that she "sustained neck and right wrist injuries when she fell on wet woodchips while going to the parking lot" (WC Petition at 1). Both Plaintiff and her counsel signed the Workers' Compensation Petition. In doing so, they understood and acknowledged that "ANY PERSON WHO, KNOWINGLY AND WITH INTENT TO INJURE, DEFRAUD, OR DECEIVE ANY EMPLOYER, OR EMPLOYEE, INSURANCE COMPANY OR SELF-INSURED PROGRAM, FILES A STATEMENT OF CLAIM CONTAINING ANY FALSE OR MISLEADING INFORMATION IS GUILTY OF A FELONY OF THE THIRD DEGREE" (WC Petition at 3). The Court takes judicial notice of the Workers' Compensation Petition and other filings in Plaintiff's workers' compensation case pursuant to FED. R.EVID. 201.
[11] The workers' compensation settlement was executed approximately five months after the state court entered the consent judgment in the underlying tort action (Doc. 18-2). However, The Hartford and Plaintiff agreed that: "This settlement agreement ... shall not act as an election of remedies or settlement as to Ms. Sinni's civil suit against Absolutely Massage, Inc. and Aaron Cullen ... and Ms. [sic] Sinni reserves all rights to suit against said company and individual, +/or [sic] as to assignment + [sic] suit against Scottsdale Insurance / [sic] Indemnity Co" (Doc. 51-43 at 2). On October 15, 2008, Judge Condry approved the allocation of Plaintiff's workers' compensation settlement proceeds. See Order Granting Motion for Approval of Allocation of Settlement Proceeds filed on 10/15/2008, available at http://www.jcc.state. fl.us/JCC/searchJCC/, Case No. 06-000191WJC (Condry, J.)
[12] The Association for Cooperative Operations Research and Development (ACORD) is apparently a non-profit standards organization that, inter alia, has developed standardized claim forms that are used in the insurance industry. See generally About Acord, http://www.acord.org/about.
[13] Scottsdale received the letter and "Notice of Occurrence" on or about June 22, 2006 (Doc. 53-20 at 4, Stutheit Dep. at 11).
[14] Plaintiff suggests that on May 11, 2007, her counsel sent a letter and a copy of the complaint to the independent adjuster whom Scottsdale had previously assigned to investigate Plaintiff's "Notice of Occurrence" (Doc. 51-22 at 2). There is no evidence, however, that the independent adjuster ever forwarded the complaint to Scottsdale. Indeed, the affidavit of Plaintiff's counsel that addresses whether he actually enclosed a copy of the complaint with his May 11, 2007 letter is neither signed nor notarized (see Doc. 51-91).
[15] During that same telephone conversation, but apparently subsequent to Scottsdale's offer to defend, Mr. Truesdell supposedly told the adjuster that his clients had entered into the Coblentz agreement (Doc. 53-28 at 35). Neither party has addressed the admissibility of this hearsay statement. Even assuming, however, that the statement and purported sequence of events were true, they are irrelevant. Accordingly, the Court has not considered or relied on this evidence.
[16] The policy obligated the insureds to provide notice of the claim to Scottsdale "as soon as practicable." The insureds clearly breached this provision of the policy.
[17] It is worth noting that the parties to a Coblentz agreement have an incentive to move quickly in order to catch the insurer in a Coblentz waiver trap, as discussed, infra.
[18] While the Court is reluctant to address issues that have not been briefed by the parties, the dicta in Wright is also problematic on other grounds. First, in Indian Harbor Ins. Co., 998 So.2d at 679, the Fourth District Court appears to have implicitly overruled the dicta in Wright. Second, a rule denying insurers the opportunity to raise policy exclusions and contest their duty to indemnify on the meritsat what is likely to be the only adversarial stage in the course of otherwise non-adversarial "ligation"could raise significant due process concerns. See, e.g., Hansberry v. Lee, 311 U.S. 32, 40, 61 S.Ct. 115, 85 L.Ed. 22 (1940) (courts have a duty to examine the course of litigation and ascertain whether litigants whose rights have been adjudicated were accorded due process) (citing W. Life Indem. Co. of Ill. v. Rupp, 235 U.S. 261, 273, 35 S.Ct. 37, 59 L.Ed. 220 (1914)); Scull v. State, 569 So.2d 1251, 1252 (Fla. 1990) ("Due process envisions a law that hears before it condemns, proceeds upon inquiry, and renders judgment only after proper consideration of issues advanced by adversarial parties") (citations omitted).
| null |
minipile
|
NaturalLanguage
|
mit
| null |
I went looking for the Key Bridge good Samaritan. I didn’t find one. I found many.
Last week, I wrote about a man who in February jumped off the Key Bridge. He survived the suicide attempt and is getting good treatment for the mental illness that drove him to such a desperate act. His father wanted to thank the mystery passerby who he was told had grabbed a boat and rescued his son from the freezing water.
John Kelly writes "John Kelly's Washington," a daily look at Washington's less-famous side. Born in Washington, John started at The Post in 1989 as deputy editor in the Weekend section. View Archive
Many witnesses contacted me and said they are near certain the man swam out of the water himself. (The man himself has only the fuzziest recollection.)
Though an Arlington County Fire Department spokeswoman told me that someone on a running trail — “an adult male in the military” — had grabbed a kayak, she did say that since the EMT crew was busily involved in patient care, it’s possible the story had become garbled.
The D.C. police report about the incident said the man swam to the shore unaided. I think this in no way diminishes what happened that morning.
Anne Hancock holds Arrow, a cat who was found in January wandering near Ballston Common Mall and taken to the Animal Welfare League of Arlington. He was blind and had been shot multiple times with a BB gun or shotgun. He was adopted by Anne Hancock of Arlington. X-rays reveal the extent of the injuries. The pellets will be left in. (Courtesy Animal Welfare League of Arlington)
I heard from three people who were on the bridge, two in cars, the other walking on the sidewalk. All expressed concern for the man.
I heard from Caleb Jackson, too. The retired D.C. police officer from Upper Marlboro was between takes of an independent Web series that was shooting near a boathouse when he happened to look up and see the man unhesitatingly climb the fence on the Key Bridge and jump off. Caleb rushed toward the water, stopping only to call 911.
When the man emerged, dripping wet, Caleb bundled him in a blanket and led him to his warm SUV to await the ambulance.
A doctor who happened to be running by stopped to check the man’s condition, Caleb said.
“It seemed like everybody who saw him was concerned,” Caleb said. “Everybody tried to do their part.”
The man’s father said he may never know exactly what happened that morning. But, he said, “the people I’ve talked to indicated there were a lot of people willing to help.”
That’s encouraging. So is the fact that his son was so quick to get out of the water — that he didn’t stay there, succumbing to hypothermia. I hope that bodes well for his future.
The scene that frigid February morning must have seemed cinematic: traffic on the bridge, joggers below, pedestrians everywhere and then, suddenly, a falling man.
Elisabeth Blaug and her husband saw it happen, and she told me that she was haunted by it, breaking into tears whenever the memory came back. That’s why she was so glad to read that the man was doing well.
“I know that this man comes from a loving and caring family that will do everything it can to help their beloved son,” she wrote. “Please let the family know that two strangers have held this man in their thoughts and prayers, and will continue to do so.”
This is amazing when you consider what the feline has been through. He was discovered in January near Ballston Common Mall. At the time, he was feral and in rough shape. He was undernourished. One eye was shriveled. The other had a detached retina. He was taken to the Animal Welfare League of Arlington, where he was X-rayed to check for possible lung problems.
That’s when the vet found someone had used Arrow for target practice. His body was peppered with more than 80 pellets, fired from either a BB gun or a shotgun.
Anne had no qualms about adding a blind, shot-ridden Arrow to her existing menagerie: cats Brianna and Willie.
Arrow is slowly acclimating to Anne’s Arlington home. She’s confined him to one floor of the house but will slowly introduce him to the rest of it. “He wanders from room to room, sensing where he is,” she said.
A happy ending to a sad situation, I pointed out.
“It’s very sad,” Anne said. “It’s worse than sad. It’s criminal to treat any living creature so inhumanely. It’s amazing that he is so trusting and affectionate in spite of it. . . . I think when these cats find a safe haven, they don’t even care about the food. They just want to be loved and cared for. They’re very grateful.
“I think I’m luckier than he is. He really is a wonderful, super special cat.”
Comments our editors find particularly useful or relevant are displayed in Top Comments, as are comments by users with these badges: . Replies to those posts appear here, as well as posts by staff writers.
To pause and restart automatic updates, click "Live" or "Paused". If paused, you'll be notified of the number of additional comments that have come in.
Comments our editors find particularly useful or relevant are displayed in Top Comments, as are comments by users with these badges: . Replies to those posts appear here, as well as posts by staff writers.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Remixxing the mainstream news one blog post at a time from the shores of Venice Beach. News, politics & conspiracy theories about world issues. All posts are opinions meant to foster comment, reporting, teaching & study under the "fair use doctrine" in Sec. 107 of U.S. Code Title 17. No statement of fact is made or should be implied. Ads appearing on this blog are solely the product of Blogger.com and do not necessarily reflect the opinions of Remixx World!
Monday, August 17, 2009
Speaker Pelosi, I believe this above videos of fake ACORN protesters getting booed while leaving Sen. Specter's town hall meeting is the definition of "astroturf" as you stated. I must thank you, because I never knew what the word meant until you said it, so you helped me increase my knowledge. Nevertheless, the local Pennsylvania people booed ACORN and SEIU members who were bussed into a recent Sen. Specter town hall in Pennsylvania.
The American people have to ensure that the genuine grassroots anger happening all over the country is not hijacked by Republicans and Democrats to further their two-party paradigm agenda that divides the country. No matter how much the Republicans and Democrats debate, the only guarantees are a bigger government and a ballooning debt.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Weekly Blog
I sang to myself as I folded laundry yesterday. With my sister at school and my parents at a chiropractor's appointment, I had the house to myself. And it was a little scary.I normally enjoy being alone. My room is my kingdom. It's nice and quiet, decorated in gentle greens with splashes of brown and muted pink. I have a yarrow plant on the windowsill, a Bible on the nightstand, and my precious laptop reigns from her place of honor on my desk. But when the entire house is mine? I thought I heard someone flip a light switch and nearly had a conniption fit.I'll be eighteen in a month. This is ridiculous.If I'd lived a few hundred years ago, I'd probably be married by now, or at the very least I'd be looking for a suitor. I'd know everything I needed to know in order to run a house. I'd be able to cook more than tacos, pizza, and hamburgers--and that without asking Mom if the meat's fully cooked. Heck, I'd be considered a full-fledged adult. An empty house wouldn't be scary at all. In fact, it would mean that I'd be able to get some work done in peace. But for me...oh, there's still an eight-year-old inside me who jumps when the window creaks.
But today? I'm just now starting down the road to independence. I have a job, but my mom waits up until I get home. I have chores, but if I forget about them, it's not a big deal. I have a car, but I got help paying for it. I have cooking skills, no matter how limited they are, but I'm still given money for fast food and an invitation to Pa and Grammy's house for lunch. I have a degree of responsibility, but I'm told that it's okay not to use it.Teenagers, I read the other day, are a fairly recent invention. Even in the early 1900s, children turned into adults without going through this purgatory of high school, peer pressure, and internet memes. My paternal grandfather was left as the man of the house as a teenager--I want to say fourteen, but I'm not sure. He quit school and started working to support his family. Today, my great-uncle Louie, Grandpa's brother, has nothing but good to say about him. But what about the standard teen of today? The ones who are so lazy they'd rather eat pre-packaged junk food than pour a bowl of cereal.I'm glad people no longer die if they don't learn basic life skills. I'm grateful for Kohl's, Aldi's, and frozen taquitos. And, after washing so many dishes at Domino's, I'm more grateful than ever for automatic dishwashers. But life skills aren't replaceable. They can't be digitized or automated, either. I think everyone should learn how to cook at least one meal, do laundry, and clean a bathroom. And everyone needs the gumption to actually do it now and again.What's your favorite thing to do when you're home alone? Let me know in the comments below! God bless you, dear reader, and don't forget to Like us on Facebook!
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Moms Aren't the Only Ones Who Struggle With the Loss of a Child
The loss of a child is never an easy thing to experience as a parent. And whether that loss occurred through miscarriage, stillbirth, or death in infancy, each experience is uniquely heartbreaking. Navigating this devastating experience can feel incredibly isolating for many parents, but often even more so for dads who are often completely left out of the conversation.
As the host of the popular podcast "The Birth Hour," Bryn Huntpalmer has seen firsthand the power that sharing stories can have for women—and the healing and empowerment that can occur as a result. But why wasn't this avenue of healing as accessible for dads, she wondered. She believed they should be able to feel like they can share their perspectives on birth and their experiences with loss as well.
"Hosting a birth stories podcast, I hear a lot of beautiful
and empowering stories but I also hear a lot of stories of loss. The fact is
that 1 in 4 pregnancies end in a miscarriage and these stories often are never
shared," shares Huntpalmer.
"There are lots of
different ways to recognize the moms of rainbow babies (babies born shortly
after the loss of another baby) and even the babies themselves, but not much
for dads. Dads are affected by a pregnancy loss too and often they suffer
silently as they are the ones charged with the job of helping the mom through
her grief."
Huntpalmer wanted to find a tangible way to help dads feel recognized in their loss and invite them into the narrative, and after much brainstorming, she finally found a way—the amazing Rainbow Dad t-shirt.
Photograph by TATTRD Photos
She explains, "I know rainbow moms are so grateful for the support of their
partners and want them to be recognized for everything they do as well as for
what they have lost. I hope that these shirts serve as a sweet gift for women
to give to the fathers of rainbow babies as well as something the men will wear
proudly."
I couldn't love this more. And dads are chiming in to share their appreciation for the inclusion the t-shirt offers.
Father and babywearing consultant Ben Winton, who will be receiving a Rainbow Dad t-shirt for Father's Day says,"When we lost our
first, I was devastated and at times I felt like my pain didn't matter. So when
our oldest son was born, I was completely over the moon, but it is a little
hard to think that he could have been our second child. The shirt brings awareness
that men do have feelings—even if people aren't always aware of them."
So, to all the rainbow dads (and moms) out there: thank you for sharing your stories. Our hearts are with you in your journey of loss and we are thrilled to celebrate those sweet rainbow babies alongside you!
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Advertising Read more
Washington (AFP)
Three British former foreign exchange traders for major banks have surrendered to US authorities to face criminal charges of market manipulation, the Justice Department announced Monday.
The men were traders for affiliates of JP Morgan Chase, Barclays, Citicorp and the Royal Bank of Scotland and appeared Monday before a federal judge in New York.
Richard Usher, Rohan Ramchandani and Christopher Ashton face a single-count indictment charging that they conspired to fix prices and rig bids for dollars and euros traded on the foreign exchange spot market.
Prosecutors claim that between December 2007 and January 2013, the trio and other unnamed co-conspirators, dubbed "the Cartel" or "the Mafia," used internet chatrooms and near-daily phone calls to conspire in fixing prices.
The conspirators allegedly also helped control prices "by refraining from entering bids/offers or trading at certain times," the Justice Department said in a statement.
US officials have taken aggressive action in recent years to prosecute interest rate and foreign exchange rate manipulation by major banks.
The banks affiliated with the former traders settled with the Justice Department in 2015, together agreeing to pay more than $2.5 billion in fines over foreign exchange manipulation.
The charges carry a maximum penalty of a $1 million fine and 10 years in prison. The fine can be increased to twice the damage suffered by victims or twice the gain derived from the crime, according to the department.
Earlier on Monday, the Federal Reserve fined French giant BNP Paribas $246 million, also over foreign exchange practices.
© 2017 AFP
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Enzymatic characterization of lipid-based drug delivery systems.
The present work introduces a simple and robust in vitro method for enzymatic characterisation of surface properties of lipid dispersions in aqueous media. The initial lipolysis rate in biorelevant media, using pancreatic lipase and a self-microemulsifying formulation (SMEDDS) containing digestible lipids as substrate, was determined. The impact of incorporating two sparingly water soluble model drugs, probucol and halofantrine, into the SMEDDS was studied. It was found that both model drugs reduced the initial rate of lipolysis compared with the vehicle, probucol having a larger effect than halofantrine. The reduction of initial lipolysis rate indicates that probucol and halofantrine are bound in the water/emulsion interface limiting the substrate availability.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Profile
Lindell Wigginton
Iowa State
Guard
6’2″, 190 pounds
6’3″ 1/2 wingspan
Projection: Late First to Early Second
Strengths
Wigginton is a freakishly athletic, scoring guard. He is a fluid athlete, with elite leaping ability; he is a highlight dunk waiting to happen. Wigginton is prolific in transition as a scorer and a passer. He has a quick first step to get by defenders. Wigginton has very nice touch and feel on his floaters. He gets to the free throw line at a very good rate and draws contact well (5.2 FTA), and has a whole back of tricks when it comes to finishing at the rim. He is excellent off the ball as well as well; his cuts are lethal.
Wigginton is a very good shooter and shot creator from all over the floor. He is ok on catch and shoot attempts, but an excellent three-point shooter off of the bounce. This season, he shot a high percentage from three (40.1%) on high volume (5.5 threes per game). Wigginton’s mid-range game is great; he uses crisp footwork and creative ball handling to make space in the mid-range to get his shot off. He gets good elevation on his shot form and has a quick release. His vision is great in transition. He doesn’t turn the ball over much (15.8 TOV%) despite his high usage rate (28.4%).
Wigginton is a very willing and capable defensive player. He has quick feet, stays in a stance for the whole possession, and has good hands. He is a competitor, and always wants to guard the opposing team’s best player. His athleticism allows him to recover when he gets beat. He is active on the glass and a good rebounder for a guard (3.7 RBPG). Wigginton is a fierce competitor, a hard worker, and will get you buckets in bunches.
Weaknesses
Wigginton’s size (6’2″, 190 pounds) and small wingspan will hold him back in the NBA. He struggles with finishing against long and athletic defenders and struggles to finish through contact. His wingspan allows players to shoot over him. His lack of core strengths allows drives with little resistance given. He is undersized to play the two but is not a true point guard. He was forced to play a lot of point due to injuries in college. His vision is unspectacular and he often neglects the easy pass in favor of a tough shot. He doesn’t elevate his teammates in the half court.
His shooting is occasionally streaky, and his mechanics can be inconsistent from shot to shot. He tries to force the issue too often, resulting in wild misses. Wigginton is prone to a lot of fouls and shoots below average at the free throw (66%) line for a primary ball handler.
Latest From FPC on SportsCastr
Best NBA Fits
Golden State Warriors
The rich keep getting richer with this pick right here. Golden State could absolutely afford to gamble on a guy like Wigginton late in the first round. Wigginton would not be asked to play impact minutes for the Warriors. He would be able to develop under an elite point guard, an excellent coaching staff, and in a low-pressure situation. All the Warriors would ask him to do would be to come off of the bench and get buckets, which he is built to do. If his point guard skills take off, he could even become a long-term option if their core four goes their separate ways.
Memphis Grizzlies
This season has been a complete and utter disaster for the Grizzlies. They have the lowest scoring offense in the NBA this season (98.7 PPG), and the third lowest three-point percentage (34.8%). The Grizzlies don’t have an elite scoring guard on the roster, and Wigginton could come in and provide some instant offense. He would learn under Mike Conley, who is one of the most effective an efficient floor generals in the NBA. Besides Andrew Harrison and Wayne Selden, the Grizzlies are devoid of young talent at the guard position. Wigginton could provide a great spark off of the bench as a sixth man, or even start some games if Conley reinjures himself or Memphis can’t buy a bucket.
New York Knicks
The Knicks have the fourth worst three-point percentage this season (34.9%). Trey Burke has been great, but can he keep up his hot scoring streak next year? The Knicks lack a great scoring guard, and Wigginton could come in and score. He has never had a dominant big man like Porzingis; Wigginton could cut off of Porzingis for easy scores all day.
Overview
Wigginton unexpectedly decided to “test the NBA draft waters” after a good freshman season; he may not declare. After an injury to Iowa State’s starting point guard, Wigginton took a lot of minutes at the lead guard role. Although he struggled to be an effective playmaker, the challenge granted good experience for the young guard. His combination of small size and lack of point guard skills will hurt him against the length and athleticism of the NBA, but he will be able to score effectively day one. His scoring ability, toughness, and athleticism could have him an NBA home as early as the late first round. Lindell Wigginton is a major sleeper, a player you must keep your eye on as the draft nears.
Ben Pfeifer is the Managing Editor of the Colts for Full Press Coverage, the AFC South Division Editor, and head NBA editor. Want to continue the discussion? Contact Ben Pfeifer on Twitter @Ben_Pfeifer_, @FPC_NBA and @FPC_Colts.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Scandinavian clinical practice guideline on choice of fluid in resuscitation of critically ill patients with acute circulatory failure.
The task force on Acute Circulatory Failure of the Scandinavian Society of Anaesthesiology and Intensive Care Medicine produced this guideline with recommendations concerning the use of crystalloid vs. colloid solutions in adult critically ill patients with acute circulatory failure. Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology was used to grade the quality of evidence and to determine the strengths of the recommendations. As efficacy and harm may vary in different subpopulations of patients with acute circulatory failure, we produced recommendations for general intensive care unit (ICU) patients and those with sepsis, trauma and burn injury. For general ICU patients and those with sepsis, we recommend using crystalloids for resuscitation rather than hydroxyethyl starch and we suggest using crystalloids rather than gelatin and albumin. For patients with trauma we recommend to use crystalloids for resuscitation rather than colloid solutions. For patients with burn injury we provide no recommendations as there are very limited data from randomised trials on fluid resuscitation in this patient population. We recommend using crystalloid solutions rather than colloid solutions for resuscitation in the majority of critically ill patients with acute circulatory failure.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
All relevant data are within the paper and its Supporting Information files.
Introduction {#sec005}
============
A large cohort study showed that diabetes was an independent risk factor for chronic liver disease (CLD) and hepatocellular carcinoma (HCC) \[[@pone.0195028.ref001]\]. On the other hand, CLD is one of the major causes of death in diabetic patients \[[@pone.0195028.ref002]--[@pone.0195028.ref003]\]. These results point to a close relationship between CLD and diabetes. The liver plays a central role in glucose metabolism. Almost all patients with cirrhosis are insulin-resistant, 60% to 80% are glucose intolerant, and about 20% develop diabetes \[[@pone.0195028.ref004]\]. It is important to maintain a good glycemic control status because patients with CLD and inadequate blood glucose control have poor prognosis \[[@pone.0195028.ref005]\]. Insulin resistance and hyperinsulinemia are metabolic features of patients with CLD. Furthermore, postprandial hyperglycemia is seen frequently in patients with CLD. On the other hand, the metabolic profiles of patients with liver cirrhosis (LC) after an overnight fast resemble those found in normal humans after 2--3 days of starvation. This phenomenon reflects the fact that hepatic glycogen store is decreased in patients with LC \[[@pone.0195028.ref006]\]. For these reasons, nocturnal hypoglycemia sometimes occurs in patients with LC.
The main feature of blood glucose dynamics in patients with CLD is marked blood glucose fluctuations, such as postprandial hyperglycemia and nocturnal hypoglycemia. However, the detail of blood glucose dynamics is not well known. Glycated hemoglobin A1c (HbA1c) is the gold standard for monitoring blood glucose control in diabetes. However, HbA1c does not properly represent the glycemic control status in patients with LC due to the short lifespan of erythrocytes caused by hypersplenism \[[@pone.0195028.ref007]--[@pone.0195028.ref009]\]. HbA1c has been shown to be apparently lower in relation to hyperglycemia in patients with CLD. On the other hand, although glycated albumin (GA) is not influenced by disorders of hemoglobin metabolism, it is affected by disorders of albumin metabolism. GA level is apparently higher in relation to hyperglycemia in CLD patients due to the prolonged half-life of serum albumin caused by reduced capacity for albumin synthesis \[[@pone.0195028.ref010]--[@pone.0195028.ref011]\]. Moreover, diabetic patients with similar HbA1c levels may differ in terms of glucose stability. In other words, since HbA1c is an integrated measure of overall glucose exposure, similar HbA1c levels can be generated by different glucose profiles \[[@pone.0195028.ref012]\]. Thus, it is difficult to monitor glycemic control status accurately in patients with CLD, because none of the known markers precisely reflects that status.
Frequent blood testing, such as self-monitoring of blood glucose (SMBG) provides diabetics with accurate and discrete blood glucose levels. However, SMBG does not provide trend information on glucose, nor reflects glycemic fluctuations. Nocturnal hypoglycemia is never captured by SMBG. On the other hand, continuous glucose monitoring (CGM) systems can provide information about glucose levels every few minutes, allowing patients to view a graph of glucose levels. Furthermore, the CGM systems provide maximum information about fluctuations in blood glucose that various parameters of glycemic variability \[mean blood glucose (MBG), the difference between highest and lowest blood glucose (ΔBG), and mean amplitude of glycemic excursion (MAGE)\] by continuous analysis of interstitial glucose level throughout the day, and the device is currently used to educate patients with diabetes. In Japan, the device was covered by Medicare by the Ministry of Health, Labour and Welfare in 2009.
It is known that blood glucose fluctuations are hidden in patients with CLD. Moreover, postprandial hyperglycemia and nocturnal hypoglycemia are features of blood glucose kinetics of CLD patients. The CGM systems can help identify nocturnal hypoglycemia and prevent it from being unwanted postprandial hyperglycemia with minimal invasiveness. The results of CGM may contribute to more appropriate therapeutic strategy for blood glucose control in CLD patients with type 2 diabetes mellitus (T2DM). The aim of this study was to analyze blood glucose fluctuations using the CGM in CLD patients with T2DM.
Materials and methods {#sec006}
=====================
Patients {#sec007}
--------
The study protocol was approved by the Human Ethics Review Committee of Hiroshima University and a signed consent form was obtained from each subject. It complies with the Treaty of Helsinki. Between September 2013 and August 2015, 105 CLD patients with T2DM were enrolled in this retrospective study. The study subjects were patients with CLD diagnosed with T2DM according to Japan Diabetes Society guidelines \[[@pone.0195028.ref013]\]. T2DM was diagnosed based on the presence of one of the following criteria: (i) fasting plasma glucose level of ≥ 126 mg/dL; (ii) 2-h plasma glucose level of ≥ 200 mg/dL after 75-g oral glucose tolerance test (OGTT); or (iii) casual plasma glucose level of ≥ 200 mg/dL. According to the current revision, in addition to the above listed plasma glucose levels, HbA1c has been given a more prominent position as one of the diagnostic criteria. That is, (iv) HbA1c level of ≥ 6.5% is currently considered to indicate T2DM. All patients underwent CGM during hospitalization. Diet, including calories, was set for each individual patient during hospitalization. The selected median calorie per day was 1600 kcal (range, 1200--2000 kcal). The calorie intake and treatment contents were fixed during CGM.
Diagnosis of chronic hepatitis and liver cirrhosis {#sec008}
--------------------------------------------------
Patients were divided clinically into chronic hepatitis (CH) group (n = 51) and liver cirrhosis (LC) group (n = 54). LC was determined radiologically or by histopathological examination of liver biopsy material. Patients who could not be diagnosed radiologically or by histopathological examination, an APRI (AST to platelet ratio index) score of \> 1 was used to indicate LC. The APRI score was calculated using Wai's formula: \[(AST/upper limit of normal)/platelet count (expressed as platelets × 10^9^/l) × 100\]. The reported sensitivity, specificity, and positive and negative predictive values of APRI score \> 1 for LC are 89%, 75%, 38%, and 98%, respectively \[[@pone.0195028.ref014]\].
Measurement of CGM {#sec009}
------------------
Subcutaneous interstitial glucose levels were monitored on an ambulatory basis over 72 consecutive hours by using the CGM systems (iPro2, Medtronic, Minneapolis, MN). All patients received detailed information about the benefits, examination procedure, and possible complications of the CGM. The sensor was inserted into the subcutaneous tissue of the abdomen and removed after 96 hours, with a daily record containing 288 logging continuous sensor values. In all patients, the CGM was calibrated at least 4 times per day with a SMBG device (Medisafe-Mini, Terumo, Japan). According to the operating guidelines, the CGM was installed to monitor glucose levels in the interstitial fluid. The following parameters were computed from the recording; MBG, ΔBG, MAGE, the standard deviation of blood glucose (SDBG), area under the curve of blood glucose above 140 mg/dl, (AUCgluc ≥ 140) and area under the curve of blood glucose below 70 mg/dl (AUCgluc \< 70). MAGE quantified major swings in blood glucose but exclude minor fluctuations when assessing intra-day glycemic variability. MAGE represented the arithmetic mean difference between consecutive blood glucose peaks and nadirs when differences were \> 1 standard deviation of the mean glucose value in the same 24 hours period \[[@pone.0195028.ref015]--[@pone.0195028.ref016]\]. It is reported that the mean value of MAGE of early screening-diagnosed T2DM patients based on OGTT was 77.4 mg/dL \[[@pone.0195028.ref017]\]. In the present study, 200 mg/dL, 70 mg/dL, and 77.4 mg/dL were defined as the cutoff values for maximum blood glucose, lowest blood glucose, and MAGE, respectively. As for MBG and ΔBG, the mean values of the subgroup were defined as the cutoff values.
Comparison of glycemic parameters obtained from CGM {#sec010}
---------------------------------------------------
The level of HbA1c is commonly used as reliable index of glycemic control in diabetic patients \[[@pone.0195028.ref018]--[@pone.0195028.ref019]\]. HbA1c levels of \< 7.0% is currently considered the treatment target with respect to diabetic complications \[[@pone.0195028.ref020]\]. However, in patients with CLD, the level of HbA1c shows lower values relative to the degree of glycemia because of anemia caused by hypersplenism-associated shorter erythrocyte lifespan. Thus, in anemic patients with CLD, the level of HbA1c is not reliable. Therefore, we divided the patients into three groups; CLD patients with HbA1c levels of ≥ 7.0% (n = 64), non-anemic CLD patients with HbA1c levels of \< 7.0% (n = 29) and anemic CLD patients with HbA1c levels of \< 7.0% (n = 12). In this study, we defined anemia as hemoglobin (Hb) levels of ≤ 12 g/l in men and ≤ 10 g/l in women. Because the number of anemic CLD patients with HbA1c levels of \< 7.0% was few, we investigated the former two groups. The patients with HbA1c levels of ≥ 7.0% consisted of 33 CH and 31 LC patients, and the non-anemic patients with HbA1c levels of \< 7.0% consisted of 14 CH and 15 LC patients. Blood glucose fluctuations, such as maximum blood glucose, lowest blood glucose, MBG, ΔBG, MAGE, SDBG, AUCgluc ≥ 140 and AUCgluc \< 70 were compared between CH patients and LC patients.
Statistical analysis {#sec011}
--------------------
All statistical analyses were performed using SPSS for Windows 20.0 (SPSS Inc, Chicago, IL). Results are presented as frequencies and percentages for categorical variables and median (range) for continuous variables. Differences between two groups were assessed by using the chi-square and unpaired *t* tests. To ascertain the independent contribution of MAGE and mean glucose level, multivariate regression analysis was used. To assess the trend of MBG, ΔBG, MAGE, SDBG, AUCgluc ≥ 140 and AUCgluc \< 70 in the sample, each parameter was compared across the three groups divided by liver functional reserve using a Jonckheere-Terpstra test. *P* value of \< 0.05 was considered statistically significant.
Results {#sec012}
=======
Patient characteristics and laboratory data {#sec013}
-------------------------------------------
[Table 1](#pone.0195028.t001){ref-type="table"} shows the characteristics of the patients \[62 men and 43 women, median age: 68 years (range, 24--84)\]. There were 51 CH patients and 54 LC patients. According to the Child-Pugh classification, 31 patients were classified as class A, 18 as class B and 5 as class C. Glucose-lowering agents were already prescribed in 87 patients, and those were dipeptidyl peptidase-4 inhibitors (DPP-4I, n = 59), biguanides (BG, n = 29), sulfonylureas (SU, n = 28), α-glucosidase inhibitors (α-GI, n = 28), insulin therapy (n = 27), thiazolidinediones (n = 5), rapid-acting insulin secretagogues (n = 2) and sodium glucose cotransporter 2 inhibitors (SGLT2I, n = 1), with some patients taking more than one type of these agents. There were not significant difference blood glucose fluctuations between these glucose-lowering agents. Age, etiology, sex, BMI, and the rates of lifestyle-related diseases (hypertension, dyslipidemia and hyperuricemia) were not significantly different between CH group and LC group. The levels of AST (*P* = 0.001) and ALT (*P* = 0.049), the rates of HCC (*P* = 0.004) and APRI (*P* \< 0.001) were significantly higher in LC group than in CH group. Hb levels, platelet count, and albumin levels were significantly higher in CH group than in LC group (*P* \< 0.001). On the other hand, HbA1c, fasting plasma glucose (FPG), fasting immunoreactive insulin (FIRI) levels, and homeostasis model assessment for insulin resistance (HOMA-IR) were similar in the two groups. HOMA-IR, a method used to quantify insulin resistance, was calculated as follows: HOMA-IR index = FPG (mg/dL) × FIRI (mU/L) / 405. The rate of use of glucose-lowering agents except for biguanides were similar in the two groups. This is because biguanides is contraindicated in patients with severe hepatic impairment.
10.1371/journal.pone.0195028.t001
###### Characteristics of patients.
{#pone.0195028.t001g}
ALL patients Patients with HbA1c ≥7.0% Non-anemic patients with HbA1c \<7.0%
--------------------------------------- ------------------- --------------------------- --------------------------------------- --------- ------------------- ------------------- --------- ------------------- ------------------- ---------
**Number of patients** 105 51 54 33 31 14 15
**Age (years)** 68 (24--84) 66 (34--81) 68.5 (24--84) 0.099 66 (34--80) 69 (56--84) 0.005 68 (59--81) 68 (24--79) 0.279
**Etiology**
**HBV/HCV** 9/30 3/9 6/21 1/5 3/9 2/4 1/10
**Alcoholic** 25 10 15 6 9 1 4
**NASH** 24 19 5 15 4 3 1
**AIH** 2 0 2 0 1 0 0
**Hemochromatosis** 1 0 1 0 1 0 0
**Cryptogenic** 16 10 6 6 4 4 2
**Sex (M/F)** 62/43 29/22 33/21 0.658 20/13 19/12 0.955 6/8 7/8 0.837
**Child Pugh classification (A/B/C)** 31/18/5 21/9/1 7/6/2
**BMI (kg/m**^**2**^**)** 25.2 (15.4--67.0) 25.2 (19.2--67.0) 24.3 (15.3--48.5) 0.249 26.7 (20.0--67.0) 25.1 (15.4--36.2) 0.126 24.7 (19.2--34.8) 25.2 (17.1--48.5) 0.593
**Hb (g/dl)** 12.8 (7.5--17.1) 13.7 (8.6--17.1) 11.9 (7.5--16.8) \<0.001 13.5 (9.5--16.8) 12.8 (8.2--15.7) 0.006 14.1 (11.5--17.1) 12.2 (10.5--16.8) 0.012
**HbA1c (%)** 7.3 (4.6--16.9) 7.3 (5.6--16.9) 7.3 (4.6--13.4) 0.178 7.8 (7.0--16.9) 8.1 (7.1--13.4) 0.669 6.6 (5.6--6.9) 6.2 (4.6--6.9) 0.035
**FPG (mg/dl)** 131 (68--298) 129 (86--267) 142 (68--298) 0.679 133 (86--267) 151 (93--298) 0.789 117.5 (96--156) 106 (68--194) 0.74
**Insulin (μIU/ml)** 13.3 (1.9--59.1) 12.4 (1.9--47.5) 14.5 (2.7--59.1) 0.616 11.8 (1.9--34.3) 14.2 (2.7--59.1) 0.129 15.5 (4.7--47.5) 18.3 (3.5--44.5) 0.584
**HOMA-IR** 4.37 (0.6--30.4) 3.95 (0.60--30.4) 4.55 (1.06--27.1) 0.603 3.95 (0.60--18.2) 4.75 (1.1--27.1) 0.188 4.08 (1.33--30.4) 4.52 (1.1--10.66) 0.441
**AST (IU/L)** 34.5 (7--172) 30 (7--82) 47 (14--172) 0.001 29 (16--82) 44 (17--159) 0.017 29.5 (7--60) 50 (14--172) 0.002
**ALT (IU/L)** 35 (5--158) 33 (9--96) 39 (5--158) 0.049 33 (9--96) 39 (10--158) 0.236 24.5 (9--70) 41 (5--138) 0.026
**Albumin (g/dl)** 3.8 (2.3--5.3) 4.2 (2.9--5.3) 3.5 (2.3--4.4) \<0.001 4.2 (2.9--4.9) 3.8 (2.4--4.4) 0.001 4.5 (3.2--5.3) 3.4 (2.3--4.4) \<0.001
**Platelet count (10**^**3**^**/μl)** 143 (28--688) 184 (85--367) 101 (28--688) \<0.001 197 (85--367) 123 (34--688) 0.005 143 (103--252) 106 (28--243) 0.006
**HCC (yes/no)** 46/59 15/36 31/23 0.004 5/28 15/16 0.004 5/9 11/4 0.042
**APRI** 0.8 (0.11--10.5) 0.54 (0.15--1.01) 1.31 (0.11--10.5) \<0.001 0.43 (0.19--1.01) 1.13 (0.11--5.79) \<0.001 0.67 (0.15--0.94) 1.31 (0.2--10.5) \<0.001
**Hypertension (yes/no)** 60/45 29/22 31/23 0.955 19/14 23/8 0.162 8/6 7/8 0.573
**Hyperlipidemia (yes/no)** 29/76 18/33 11/43 0.087 13/20 7/24 0.147 4/10 2/13 0.291
**Hyperuricemia (yes/no)** 22/83 7/44 15/39 0.077 5/28 8/23 0.29 2/12 4/11 0.361
**Medications for diabetes (yes/no)** 87/18 44/7 43/11 0.367 32/1 25/6 0.043 10/4 10/5 0.55
Data are median (range) values or number of patients.
CH: chronic hepatitis, LC: liver cirrhosis, NASH: non-alcoholic steatohepatitis, AIH: autoimmune hepatitis, BMI: body mass index, HOMA-IR: homeostasis model assessment-insulin resistance, AST: aspartate aminotransferase, ALT: alanine aminotransferase, HCC: hepatocellular carcinoma, APRI: asparate aminotransferase to platelet ratio index
CGM data according to liver functional reserve {#sec014}
----------------------------------------------
Subjects were also divided into three groups according to liver functional reserve (CH group, Child A group and Child B and C group). FPG levels and HOMA-IR did not show any significant difference between the three groups. On the other hand, HbA1c levels were significantly lower in Child B and C group than in CH group (*P* = 0.039). However, the levels of MBG, ΔBG, MAGE, SDBG and AUCgluc ≥ 140 increased steadily from CH group to Child B and C group in a stepwise manner ([Fig 1](#pone.0195028.g001){ref-type="fig"}). The tendency was significant for MBG (*P* \< 0.001), ΔBG (*P* = 0.005), SDBG (*P* = 0.008) and AUCgluc ≥ 140 (*P* \< 0.001). There was no significant difference in AUCgluc \< 70 due to differences in hepatic functional reserve. The proportion of patients taking biguanide was significantly higher in CH group than in the other groups (p = 0.009). Therefore, we further performed multivariate ordinal logistic regression analysis adjusting for biguanide. Among factors which was significantly associated with hepatic functional reserve in univariate analysis, MBG, ΔBG and AUCgluc ≥ 140 were still significant after adjusting for biguanide (P \< 0.001, P = 0.033, and P = 0.003, respectively), whereas SDBG was not (P = 0.115).
{#pone.0195028.g001}
Comparison of CGM parameters between CH group and LC group in patients with HbA1c levels of ≥ 7.0% {#sec015}
--------------------------------------------------------------------------------------------------
Next, we compared CGM glycemic parameters between CH group and LC group in those patients with HbA1c levels of ≥ 7.0%. [Table 1](#pone.0195028.t001){ref-type="table"} summarizes the patients' baseline characteristics and laboratory data. This subgroup consisted of 33 CH patients and 31 LC patients. Age (*P* = 0.005), AST (*P* = 0.017), rate of HCC (*P* = 0.004) and APRI (*P* \< 0.001) were significantly higher in LC group than in CH group. On the other hand, Hb levels (*P* = 0.006), platelet count (*P* = 0.005), albumin levels (*P* = 0.001) and the rate of use of glucose-lowering agents (*P* = 0.043) were significantly lower in LC group than in CH group. However, HbA1c, FPG, FIRI and HOMA-IR were not significantly different.
We compared various parameters derived from the CGM between CH group and LC group ([Fig 2](#pone.0195028.g002){ref-type="fig"}). ΔBG and MAGE were not different between CH group and LC group. On the other hand, the percentage of LC patients with MBG levels of ≥165 mg/dL was significantly higher than that of CH patients (*P* = 0.041).
{#pone.0195028.g002}
Next, univariate and multivariate analyses were performed to identify factors that contributed to the increase in mean glucose level in HbA1c levels of ≥ 7.0% patients ([Table 2](#pone.0195028.t002){ref-type="table"}). Multivariate analysis identified LC as the only factor that was independently associated with the increase in mean glucose level (odds ratio = 2.908, 95% confidence interval = 1.031--8.204, *P* = 0.044).
10.1371/journal.pone.0195028.t002
###### Results of multiple regression analysis to identify factors contributing.
{#pone.0195028.t002g}
to identify factors contributing to increased mean glucose level in HbA1c ≥7.0 patients. to identify factors contributing to increased mean glucose level in non-anemic patients with HbA1c\<7.0 to identify factors contributing to the increase of MAGE in non-anemic patients with HbA1c\<7.0
---------------------------------- ------------------------------------------------------------------------------------------ --------------------------------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------- ------- ---------------------- ------ ------ ------------------------- -------
**Age \>70 years** 0.712 0.43 0.25
**Male** 0.66 0.445 0.74
**BMI \>25 kg/m**^**2**^ 0.539 0.705 0.98
**AST \>34 IU/L** 0.961 0.127 0.23
**ALT \>43 IU/L** 0.375 0.058 N.S. 0.6
**HOMA-IR \>2.5** 0.415 1 0.09 14.716 (1.177--184.023) 0.037
**LC** 0.041 2.908 (1.031--8.204) 0.044 0.023 6.25 (1.213--32.214) 0.03 0.02 13.731 (1.334--141.326) 0.028
**HCV** 0.42 0.256 0.14
**HBV** 0.617 0.541 0.44
**alcohol** 0.844 0.139 0.43
**NASH** 0.47 0.617 0.07 N.S.
**Hypertension (yes)** 0.57 0.45 0.71
**Dyslipidemia (yes)** 0.382 0.082 N.S. 0.6
**Hyperuricemia (yes)** 0.217 0.163 0.57
**Treatment for diabetes (yes)** 0.61 0.5 0.57
For abbreviations, see [Table 1](#pone.0195028.t001){ref-type="table"}
Comparison of CGM parameters between CH group and LC group in patients with HbA1c levels of \< 7.0% {#sec016}
---------------------------------------------------------------------------------------------------
This subgroup consisted of 14 CH patients and 15 LC patients ([Table 1](#pone.0195028.t001){ref-type="table"}). The levels of AST (*P* = 0.002) and ALT (*P* = 0.026), the rate of HCC (*P* = 0.042), APRI (*P* \< 0.001), and the rate of patients on BCAA supplementation (*P* = 0.014) were significantly higher in LC group than in CH group. On the other hand, Hb levels (*P* = 0.012), platelet count (*P* = 0.006), albumin (*P* \< 0.001) and HbA1c levels (*P* = 0.035) were significantly lower in LC group than in CH group. However, FPG, FIRI, HOMA-IR and the rate of glucose-lowering agents use were similar in the both groups.
We also compared various parameters obtained from CGM between CH group and LC group ([Fig 3](#pone.0195028.g003){ref-type="fig"}). ΔBG was not significantly different between CH group and LC group. On the other hand, the percentage of LC patients with MBG levels of ≥ 145 mg/dL was significantly higher than that of the CH patients (*P* = 0.023). In the same way, the percentage of LC patients with MAGE of ≥ 77.4 mg/dL was significantly higher than that of CH patients (*P* = 0.024). HbA1c levels were significantly lower in LC group than in CH group. Nevertheless, MBG and MAGE of LC group were rather higher than those of CH group.
{#pone.0195028.g003}
Univariate and multivariate logistic regression analyses were performed to identify factors that contributed to the increase in MBG levels in non-anemic patients with HbA1c levels of \< 7.0% ([Table 2](#pone.0195028.t002){ref-type="table"}). LC was the sole factor that was independently associated with the increase of MBG levels in multivariate analysis (odds ratio = 6.250, 95% confidence interval = 1.213--32.214, *P* = 0.028). Next, we performed univariate and multivariate analyses to identify factors that contributed to the increase in MAGE in non-anemic patients with HbA1c levels of \< 7.0% ([Table 2](#pone.0195028.t002){ref-type="table"}). HOMA-IR (odds ratio = 14.716, 95% confidence interval = 1.177--184.023, *P* = 0.037) and LC (odds ratio = 13.731, 95% confidence interval = 1.334--141.326, *P* = 0.028) were identified as independent factors associated with the increase in MAGE.
Postprandial hyperglycemia and nocturnal hypoglycemia {#sec017}
-----------------------------------------------------
As previously mentioned, AUCgluc ≥ 140 increased significantly in proportion with worsening of hepatic functional reserve. However, there was no significant difference in AUCgluc \< 70 due to differences in hepatic functional reserve. Therefore, we examined more about postprandial hyperglycemia and nocturnal hypoglycemia. Postprandial hyperglycemia was defined as maximum blood glucose of ≥ 200 mg/dL, while nocturnal hypoglycemia represented minimum blood glucose of ≤ 70 mg/dL during the night. Although many of the cases were being treated with glucose-lowering agents, postprandial hyperglycemia was noted in 92% of the patients. Nocturnal hypoglycemia was recorded in 22% of patients. Furthermore, postprandial hyperglycemia and nocturnal hypoglycemia were investigated in CLD patients with HbA1c levels of ≥ 7.0% and non-anemic CLD patients with HbA1c levels of \< 7.0%. In both subgroups, the percentage of postprandial hyperglycemia or nocturnal hypoglycemia was not different between CH group and LC group ([Table 3](#pone.0195028.t003){ref-type="table"}). Nocturnal hypoglycemia was seen not only in LC patients but also in CH patients. Postprandial hyperglycemia was seen at a high rate even in patients with good HbA1c levels. We could obtain these findings only by the CGM.
10.1371/journal.pone.0195028.t003
###### The proportion of patients with postprandial hyperglycemia and nocturnal hypoglycemia.
{#pone.0195028.t003g}
n Postprandial hyperglycemia(%) Nocturnal hypoglycemia(%)
----------------------------------------- -------- ---- ------------------------------- ---------------------------
**All patients** **CH** 51 88 28
**LC** 54 96 16
**all** 105 92 22
**Patients with HbA1c of ≥7.0%** **CH** 33 88 27
**LC** 31 97 13
**all** 64 92 20
**Non-anemic patients with HbA1c\<7.0** **CH** 14 86 36
**LC** 15 93 33
**all** 29 90 34
For abbreviations, see [Table 1](#pone.0195028.t001){ref-type="table"}
Discussion {#sec018}
==========
The present study estimated the glycemic parameters in CLD patients with T2DM using the CGM systems. The CGM enables to evaluate short-term glycemic variability which cannot been captured by HbA1c levels, and thus, we can easily detect postprandial hyperglycemia and nocturnal hypoglycemia. Short-term glycemic variability was assessed by measuring MAGE of all readings during the CGM. Analysis of data of all cases showed no differences in various markers of T2DM, including HbA1c levels, between CH group and LC group, whereas MBG, ΔBG and AUCgluc ≥ 140 increased significantly in proportion with worsening of hepatic functional reserve. These results indicate worsening of glucose intolerance and insulin resistance with deterioration of hepatic functional reserve. High rates of nocturnal hypoglycemia and postprandial hyperglycemia were seen irrespective of HbA1c levels and liver pathology (i.e., CH or LC). Nocturnal hypoglycemia and postprandial hyperglycemia seemed to be features of CLD patients. Among patients with HbA1c levels of ≥ 7.0%, the percentage of LC patients with MBG levels of ≥ 165 mg/dL was significantly higher than that of CH patients. Among patients with HbA1c levels of \< 7.0%, the percentage of LC patients with MBG levels of ≥ 145 mg/dL was significantly higher than that of CH patients. Likewise, the percentage of LC patients with MAGE of ≥ 77.4 mg/dL was significantly higher than that of CH patients. Thus, regardless of the HbA1c levels, glycemic parameters derived from CGM were much worse in LC group than in CH group. In summary, CGM analysis of CLD patients with T2DM demonstrated aggravation of glycemic variability (as reflected by MAGE, MBG, ΔBG etc.) with deterioration of liver functional reserve.
Insulin resistance and hyperinsulinemia are metabolic features of patients with CLD \[[@pone.0195028.ref021]\]. Hyperinsulinemia in these patients is caused by either hepatic cell damage or portal-systemic shunting because blood glucose is delivered to the liver through the portal vein. The rate at which insulin is metabolized in the liver is reduced in patients with CLD. Moreover, despite peripheral hyperinsulinemia, insulin levels in the portal veins are low in CLD patients with portal systemic shunting \[[@pone.0195028.ref022]--[@pone.0195028.ref024]\]. The frequent blood measurements have not able to reflect the various glycemic kinetics in CLD patients. The CGM systems enable to evaluate such glycemic variety in CLD patients.
Both HbA1c and GA are less reliable markers in CLD patients with T2DM. We divided the patients into two groups; those with HbA1c levels of ≥ 7.0% and the other with HbA1c levels of \< 7.0%. In the former group, various markers of T2DM, such as HbA1c, FPG, FIRI, and HOMA-IR were not significantly different between CH and LC groups. Moreover, the proportion of patients using glucose-lowering agents was significantly higher in CH group than in LC group. Nevertheless, the percentage of patients with high MBG levels was significantly higher in LC group than in CH group. Since similar HbA1c levels can be generated by different glucose profiles because HbA1c is an integrated measure of overall glucose exposure \[[@pone.0195028.ref012]\], these results suggest that among patients with comparable HbA1c levels, the actual average blood glucose or blood glucose fluctuations in LC patients with poor liver reserve were higher than those of CH patients. Multivariate analysis identified LC to be significantly associated with increased prevalence of high MBG levels. On the other hand, in CLD patients with HbA1c levels of \< 7.0%, the percentage of patients with high MBG levels or high MAGE of LC group was significantly higher than that of CH group, despite their HbA1c levels were significantly lower than those of CH group. Multivariate analysis identified LC as significantly associated with increased prevalence of high MBG levels. Moreover, by multivariate analysis, LC and the presence of insulin resistance as significantly associated with the increased prevalence of high MAGE. These results suggest that the existence of CLD patients with satisfactory glucose control markers develop large blood glucose fluctuations or have high MBG levels. Furthermore, LC and insulin resistance were significantly associated with these phenomena. In the case without anemia with normal HbA1c levels and seemingly good glucose control, we have slipped the patients in need of treatment intervention. We could not find the abnormality of such hidden blood glucose fluctuations without CGM. The results based on CGM help for the management of T2DM in CLD patients.
Although many of the cases were being treated for T2DM, postprandial hyperglycemia was seen in 90% of non-anemic patients with HbA1c levels of \< 7.0%. Nocturnal hypoglycemia was seen in 20% of patients with HbA1c levels of ≥ 7.0%, and in 34% of non-anemic patients with HbA1c levels of \< 7.0%. To identify such cases, the existing glucose metabolism markers, such as HbA1c, were not useful. Because nocturnal hypoglycemia may remain asymptomatic, it could not be recognized without CGM.
Our study has several limitations. First, we did not evaluate CLD patients without T2DM, because medical insurance do not cover for CGM in non-diabetic patients. It is possible that postprandial hyperglycemia and nocturnal hypoglycemia exist in patients with CLD who do not meet the diagnostic criteria for T2DM. We expect to be allowed to use of CGM examination in the future in CLD patients without T2DM. Second, the study did not evaluate the role of each etiology of CLD due to the small number of patients. Finally, we could not evaluate the glycemic parameters of patients without any liver disease as control group because only patients with some liver disease visit in our department.
Conclusion {#sec019}
==========
In conclusion, this present study showed that in CLD patients with T2DM, the decrease in liver functional reserve is associated with worsening of parameters of glycemic variability including MAGE, MBG, ΔBG, etc. The CGM systems were useful in finding hidden abnormalities of blood glucose fluctuations in CLD patients with T2DM, especially in non-anemic CLD patients with HbA1c levels of \< 7.0%. Furthermore, the CGM systems were useful in detecting asymptomatic nocturnal hypoglycemia in CLD patients. Information obtained from the CGM should be used to customize treatment to reduce episodes of nocturnal hypoglycemia by which to maintain the liver functional reserve.
[^1]: **Competing Interests:**The authors have declared that no competing interests exist.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Electrophysiologic effects of bepridil in normal and infarcted canine myocardium.
The electrophysiologic effects of bepridil, 10 mg/kg i.v., were determined in normal noninfarcted and in infarcted ventricular myocardium in 8 urethane-anesthetized dogs 4-6 days after anterior myocardial infarction. At drive cycle lengths of 400 and 333 ms, bepridil significantly increased relative (RRP) and effective (ERP) refractory periods in both normal ventricular tissue (mean increases, RRP 7-14%, ERP 5-6%, p less than 0.05-0.01) and in infarcted ventricular tissue (mean increases, RRP 12-15%, ERP 13-14%, p less than 0.01). Bepridil also selectively prolonged the local activation delay in infarcted ventricular myocardium (mean increases 37.5-45.1%, p less than 0.01), while ventricular excitation thresholds were not altered by bepridil in either normal or infarcted myocardium. Before bepridil administration, programmed ventricular stimulation initiated sustained ventricular tachycardias in 6 of the 8 postinfarction dogs tested. After bepridil, 2 of the 6 previously responsive animals were rendered noninducible, 3 animals responded to programmed stimulation with nonsustained tachyarrhythmias of relatively slower rates, and the one remaining dog responded with sustained ventricular tachycardia (VT). These data suggest that increases in refractoriness in both normal noninjured and in ischemically injured ventricular tissue, with a selective delay in conduction in ischemically injured tissue, contribute to the antiarrhythmic actions of bepridil in the setting of myocardial infarction.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
The statistical mechanics of dynamic pathways to self-assembly.
This review describes some important physical characteristics of the pathways (i.e., dynamical processes) by which molecular, nanoscale, and micrometer-scale self-assembly occurs. We highlight the existence of features of self-assembly pathways that are common to a wide range of physical systems, even though those systems may differ with respect to their microscopic details. We summarize some existing theoretical descriptions of self-assembly pathways and highlight areas-notably, the description of self-assembly pathways that occur far from equilibrium-that are likely to become increasingly important.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Howdy Ya'll! Everyday we will post a random picture from the great city of Houston, Texas.
Come...See...Experience Houston!
If there is anything we can do for you...snap a picture, send a postcard, swap links...just let us know!
Tuesday, March 27, 2007
Rockets Game!
This is from my lovely TV at home. I watched the Rockets play the Bucks last night...which they won 106-87. I loved the score in this picture!
0 Bouquets of Flowers:
Howdy Ya'll! Everyday we will post a random picture from the great city of Houston, Texas.
Come...See...Experience Houston!
If there is anything we can do for you...snap a picture, send a postcard, swap links...just let us know!
| null |
minipile
|
NaturalLanguage
|
mit
| null |
One week in to 2019 and the perfect time to start making predictions about what the year ahead holds for AR porn.
Many don’t know this but augmented reality porn has been around for almost a decade now, although, it’s only been the last year it really begun accelerating. Since 2010 we’ve seen a bunch of developers taking their hands to AR such as PinkVisual and EligoVision, but nothing ever really came from any of them.
Fast Forward to 2018 And we Start to See Vast Improvements.
Smartphones have become more equipped to run AR software resulting in some augmented reality porn demos and applications coming to light. Both ARKit and ARCore have improved and now offer some pretty cool features that have come in use for AR porn developers. 3D Object detection and tracking, for example have been some of the key features that allow a number of augmented reality porn apps to work as they do.
2018 saw Casey Calvert and Donny Sins become the world’s very first AR pornstars. It saw the introduction of AI, voice control and custom personalities. It saw AR hentai, 3D body scans, digitally created companions and holograms of real-life pre-recorded girls performing in augmented reality.
Last year has shown that AR porn is on the rise, which begs the questions, what can be expected in 2019? And will it be the year that it goes mainstream? These questions are dependent on a few factors.
Technology
There’re predictions all over the internet about the growth of AR tech and what can be expected over the next few years. Some sources report that the AR industry worth is gonna grow as high as $170 billion within 3 years, while others estimate it to be as low as $60 billion within the same time-frame. It’s hard to make an exact prediction of the speed in which AR grows, but what is clear is 2019 is gonna be a big year for AR technology.
Mobile
5G is set to come in to force in 2019. This is important for augmented reality porn because, to put it simply, the future of AR is dependent on a reliable 5G network. The highly complex input mechanisms of AR require a lot of data to be processed and 4G just isn’t up to the mark.
Immersion is key for AR porn so a bad connection will ruin the experience. The next generation of video formats will be much more data-intensive than what we see today and will require a much stronger network. ABI Research anticipates that 5G will bring about a 10X improvement in throughout, a 10X decrease in latency, a 100X improvement in network efficiency and a 100X improvement in traffic capacity compared to a 4G network. This enormous increase in mobile network speed, capacity, and reliability will see superfast streaming and instantaneous downloads.
Headsets
Mobile will continue to be the main driver for mainstream consumer adoption and accessibility in 2019. However, this is the year we can expect to see the release of Hololens 2. We can expect the new HoloLens to be lighter, considerably smaller and with a field of view that is twice as large as the current one. We can also expect it to include full hands tracking, eye tracking and integration with AI thanks to a dedicated chip.
While we’re just days away from the CES 2019, some are asking, is this when Microsoft will showcase the second generation Hololens? The “Out of Stock” label for the current Hololens on the Microsoft store could be an indication that the launch of the 2nd gen could be coming soon. Perhaps they will, but I don’t think so. I think it’s likely that Microsoft will launch it at their BUILD conference in April or May, this seems a more appropriate audience for such a big launch.
Rumours say the cost of the new Hololens will be between $1500-$2000. The price is less than Microsoft’s current mixed reality headset, but still too expensive to be considered consumer-friendly. Considering Magicleap has prohibited adult content on their device and Meta enters 2019 with funding issues, its currently on Microsoft’s shoulders to provide a device that can power truly immersive AR porn experiences.
Far less expensive options such as Holokit and Zapbox are available which work using your smartphone, similar to Google Cardboard but for augmented reality.
Other companies have made AR glasses such as North Focals, Vuzix Blade; Snap Spectacles v2 and Bose. However, these are products are smart specs and come in little or no use for AR porn.
So Will 2019 be The Year AR porn Becomes Mainstream?
The short answer to your question is no. The industry is now starting to pick up pace and 2019 is going to set the foundations.
We can expect more demos and updates to come in 2019. Badoink, for example, made their AR porn debut last year with the Susy Gala striptease are rumoured to release more content this year.
3D Hologroup are always updating their software and expect to see more with Naughty America’s strip club app.
With AR capabilities set to massively improve, this will likely gain the interest of other developers and porn companies so expect to see more content being released by other studios.
There’s still a road ahead before augmented reality porn really takes off, but 2019 should be an interesting year.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Luke Maye and young forwards lead North Carolina frontcourt
Share
When first-year Tony Bradley decided to enter the NBA Draft in May, he left the North Carolina men's basketball team with a gaping hole in its frontcourt.
And, though head coach Roy Williams may not like it, UNC will have to depend on a new crop of first-years to fill the roles previously held by Bradley and now-departed seniors Kennedy Meeks and Isaiah Hicks.
With Luke Maye as the sole returning forward from the rotation on last season’s national championship team, the Tar Heels will be forced to turn to first-years Garrison Brooks, Brandon Huffman and Sterling Manley to patrol the paint this season.
At his preseason press conference, Williams emphasized the importance of rebounding and solid interior play. But he expressed major concerns over the inexperience his Tar Heels face in that department.
“That’s the scary part — what’s gonna happen up front,” he said. “I still believe the single most important factor — win or lose — is rebounding.”
Editors Picks
Rebounding was a key to UNC’s success last season as the Tar Heels led the nation in both total rebounding rate and offensive rebounding percentage. However, none of last year’s top four rebounders — Hicks, Meeks, Bradley, and Justin Jackson — are returning to Chapel Hill this season. Those four players alone accounted for roughly 55 percent of the Tar Heels’ rebounds in 2016-17.
Most likely to fill this vacancy in the paint is Brooks. The four-star recruit from Lafayette, Ala., decommitted from Mississippi State in March before committing to UNC late in the recruiting process.
Forward Garrison Brooks (15) elevates for a block against East Carolina in the Smith Center on Nov. 5.
Brooks started in the Tar Heels’ first scrimmage against Barton College, tallying 13 points and nine rebounds in 19 minutes of play. ESPN national recruiting director Paul Biancardi said he thinks Brooks is the most refined and prepared of UNC’s three first-years.
“I think he can score off the block,” he said. “I think he may be the best of the three with his back to the basket.”
A more raw talent, Huffman should help shore up the Tar Heels' interior defense. The Alaska native moved to North Carolina to play his senior season at Raleigh’s Word of God Christian Academy in order to gain more exposure and committed to UNC in September of his senior year.
Huffman is still fairly unproven, and Williams expects inconsistency this season, especially if Huffman is forced to play away from the rim. But he's also flashed potential with stat lines like six points, seven rebounds and a game-high five blocks against Barton College.
“Brandon Huffman’s going to break a backboard,” Williams said. “One of the most powerful dunkers I’ve ever had, and he’s a quick jumper to get that off. But outside of three feet, it’s a little more of a question mark.”
The final first-year in the mix, Manley, will probably receive the least playing time, though he should still see meaningful action. The three-star recruit from Pickerington, Ohio, impressed in the Barton scrimmage, racking up eight points and seven rebounds in only eight minutes of play.
After bursting onto the national stage with his Elite Eight heroics against Kentucky, Luke Maye enters his junior season as the only forward with any collegiate experience. As the fourth forward in the Tar Heels’ rotation last season, the Huntersville native hauled in 15 rebounds against a top ten Florida State team that featured two seven-footers and future NBA lottery pick Jonathan Isaac. He was also named the South Region’s Most Outstanding Player in the NCAA Tournament.
Maye knows that the departures from last year mean he will be needed to step into an expanded role in his junior season.
“We have some great scorers on the perimeter,” he said, “but we still need that inside presence that we lost three big men that were really key for us last year inside.”
Plenty will change between the Tar Heels’ opening game against Northern Iowa on Friday and March, but expect the frontcourt to be a work in progress this season.
Considering their talent at the guard position, the Tar Heels may also roll out a smaller lineup occasionally – a clear departure from previous seasons’ dependence on forwards.
Williams isn’t ready to anoint any of his first-years as the next Meeks or Brice Johnson just yet, but he knows it’s a long season and expects to see improvement as the year goes on.
“If I had to play for my life (right now),” he said, “I may not play any of them — play really small. But that’s the reason you get to practice.”
“Somebody’s gotta come through and I have no idea who it’s going to be.”
| null |
minipile
|
NaturalLanguage
|
mit
| null |
1. Introduction
===============
Room temperature ionic liquids (RTILs) have been receiving considerable attention as innovative solvents for metal catalyzed organic reactions \[[@b1-ijms-12-03989]--[@b6-ijms-12-03989]\]. Most ionic liquids are based on ammonium, imidazolium, phosphonium, pyridinium, sulphonium, picolinium, pyrrolidinium, thiazolium, oxazolium, and pyrazolium cations. The important properties of ionic liquids, such as affinity to water and miscibility with other solvents, can be adjusted by the proper choice of alkyl substituents and counter anion. The most commonly used ionic liquids in olefin metathesis are 1,3-dialkylimidazolium and 1,2,3-trialkylimidazolium salts. Buijsman and co-workers \[[@b3-ijms-12-03989]\] were the first group that reported the application of ionic liquids as a medium for olefin metathesis. They studied ring closing metathesis (RCM) of several substrates using Grubbs first and second generation catalysts in RTILs, and \[bmim\]\[PF~6~\] was found to be the optimal solvent for the reactions. The activity and recyclability of Grubbs second generation catalyst in the self-cross-metathesis of some terminal olefins in RTILs has also been reported by Tang and co-workers \[[@b7-ijms-12-03989]\]. Self-metathesis of 1-octene in 1,3-dialkylimidazolium and 1,2,3-trialkylimidazolium type ionic liquids have been reported by Williams and co-workers \[[@b8-ijms-12-03989]\]. The results from the study showed that ionic liquids with shorter alkyl chain length enhanced the catalytic activity while the longer ones inhibited it. To overcome the problem associated with the leaching of products into the substrate, many charged catalyst systems have been employed \[[@b9-ijms-12-03989]--[@b14-ijms-12-03989]\].
The present work is an extension of our previous one which highlighted the self-metathesis reaction of methyl oleate in \[bmim\]\[*X*\] type ionic liquids \[[@b15-ijms-12-03989]\]. In this study, we report the self-metathesis of methyl oleate and methyl ricinoleate using the catalysts \[**1**--**4**\] in \[bmim\]\[*X*\] and \[bdmim\]\[*X*\] type ionic liquids ([Scheme 1](#f5-ijms-12-03989){ref-type="fig"}). The study also compares the metathesis activity in \[bdmim\] and \[bmim\] type ionic liquids. Catalyst recycling was also carried out in ionic liquids with the catalysts **1**--**4**.
2. Results and Discussion
=========================
2.1. Self-Metathesis of Methyl Oleate
-------------------------------------
The self-metathesis of methyl oleate was carried out in the presence of ruthenium alkylidene catalysts **1**--**4** at 40 °C. The Hoveyda-Grubbs catalysts **3** and **4** were evaluated in 1-butyl 3-methylimidazolium (bmim) and 1-butyl 2,3-dimethylimidazolium (bdmim) based ionic liquids with different anions (PF~6~^−^, BF~4~^−^ and NTf~2~^−^) ([Scheme 1](#f5-ijms-12-03989){ref-type="fig"}). [Figure 1](#f1-ijms-12-03989){ref-type="fig"} shows the activity of **3** in \[bmim\] and \[bdmim\] type ionic liquids. For catalyst **3**, similar conversions were observed in \[bmim\] and \[bdmim\] type ionic liquids (ILs). As there was not much difference in the catalytic activity, the benefit of selecting a particular IL (among PF~6~^−^, BF~4~^−^ and NTf~2~^−^) was found to be of less importance. For catalyst **4**, the highest activity was found in \[bdmim\]\[BF~4~\] with a conversion of 85% ([Figure 2](#f2-ijms-12-03989){ref-type="fig"}). Comparing the cations of the ionic liquids, the C2-methylated imidazolium cation \[bdmim\] led to a slightly better conversion. The benefit of using \[bdmim\] cation in the ethenolysis of methyl oleate has been previously reported by Thurier *et al.* \[[@b16-ijms-12-03989]\].
To study the effect of temperature on metathesis activity, the reactions were conducted at temperatures 40 °C and 60 °C. The results are summarized in [Table 1](#t1-ijms-12-03989){ref-type="table"}. An increase in substrate conversion was observed with increase in temperature. At both the temperatures, the highest substrate conversion was seen with catalyst **4**. Generally for all the runs, Hoveyda-Grubbs catalyst **4** exhibited good activity, although isomerization and formation of secondary metathesis products \[[@b17-ijms-12-03989]\] seems to be a drawback. Hoveyda-Grubbs catalyst **3** and **4** showed good solubility in ionic liquid compared to Grubbs catalyst **1** and **2**. The reason for the high activity of the Hoveyda-Grubbs second generation catalyst can be attributed to the absence of phosphine ligand. It has been proved that phosphine ligand suppresses the catalyst activity in Ru-catalyzed olefin metathesis \[[@b18-ijms-12-03989]\]. The presence of free phosphine in solution can inhibit coordination of olefins to the transition metal centre by re-association with the active Ru complex. With the non-phosphine Ru complex (catalyst **4**), activation occurs through the loss of O→Ru chelation. The styrenyl ether ligand completes less effectively with olefin substrates for Ru chelation \[[@b19-ijms-12-03989]\]. \[bdmim\] type ILs can prevent the carbene formation of catalyst, which is believed to be formed by the deprotonation of imidazolium cation or from the oxidative addition of imidazolium to Ru centre \[[@b20-ijms-12-03989]\].
2.2. Self-Metathesis of Methyl Ricinoleate
------------------------------------------
The self-metathesis of methyl ricinoleate was carried out in the presence of catalysts **1** and **2** in \[bmim\]\[X\] type ILs. [Table 2](#t2-ijms-12-03989){ref-type="table"} summarizes the activity of catalysts **1**--**4** for the self-metathesis of methyl ricinoleate. For catalyst **1**, the highest activity was shown in **5a** with a substrate conversion of 45% and the activity was found to be in the order PF~6~ \> BF~4~ \> NTf~2~ ([Figure 3](#f3-ijms-12-03989){ref-type="fig"}). For catalyst **2**, the same trend was followed, with the reaction in **5a** yielding the highest conversion of 58% ([Figure 4](#f4-ijms-12-03989){ref-type="fig"}). The primary metathesis products (PMP) obtained from the metathesis of methyl ricinoleate (**A**) were 9-octadecene- 7,12-diol (**B**) and dimethyl 9-octadecenedioate (**C**), as illustrated in [Scheme 2](#f6-ijms-12-03989){ref-type="fig"} \[[@b21-ijms-12-03989]\].
2.3. Catalyst Recycling in Ionic Liquids
----------------------------------------
As solvents, ionic liquids are more advantageous than conventional organic solvents, which make them recyclable and environmentally friendly. It has been reported that the ionic liquids could be reused for at least three runs \[[@b3-ijms-12-03989],[@b16-ijms-12-03989],[@b22-ijms-12-03989],[@b23-ijms-12-03989]\]. In a typical experiment, metathesis reaction was carried out in ionic liquids, and, after extraction of reaction products with heptane (2 × 3 mL), the glass reactor was loaded with fresh methyl oleate and was introduced for a new run.
To study the recyclability, Ru catalysts **1**--**4**, in different ionic liquids, were subjected to consecutive runs ([Table 3](#t3-ijms-12-03989){ref-type="table"}). Grubbs catalyst **1** proved to be stable in three consecutive runs, with the third run resulting in a decreased conversion. Grubbs catalyst **2** showed an increased conversion in the third cycle when compared with catalyst **1**. Comparing the Hoveyda-Grubbs catalysts **3** and **4**, the best result was obtained with catalyst **3**. The results prove that catalyst **3** can be reused for three runs without loss of activity. In spite of good activity in the first run, the second generation Hoveyda-Grubbs catalyst **4** showed low catalytic activity during the second run.
3. Experimental Section
=======================
3.1. Materials and Apparatus
----------------------------
1-Butyl-3-methylimidazolium hexafluorophosphate (\[bmim\]\[PF~6~\]), 1-butyl-3-methylimidazolium tetrafluoroborate (\[bmim\]\[BF~4~\]), 1-butyl-3-methylimidazolium bis(trifluoromethylsulphonyl)imide (\[bmim\]\[NTf~2~\]), 1-butyl-2,3-dimethylimidazolium hexafluorophosphate (\[bdmim\]\[PF~6~\]), 1-butyl-2,3- dimethylimidazolium tetrafluoroborate (\[bdmim\]\[BF~4~\]), were all reagent grade chemicals from Sigma-Aldrich. Methyl oleate (≥99%) and methyl ricinoleate (\>99%) were obtained from Sigma-Aldrich and were treated with activated alumina and stored under N~2~ atmosphere at a subzero temperature. Ethyl vinyl ether was purchased from Fluka. Nonadecane purchased from Fluka was used as the internal standard (**IS**). Ruthenium catalysts **1**--**4** were stored under N~2~ and used as purchased from Sigma-Aldrich. Chromatograms were obtained using Varian Star 3400 *CX* GC equipped with a DB-624 capillary column (J&W Scientific, 30 m × 0.53 mm) and a flame ionization detector (FID). The oven temperature was held at 200 °C and then increased to 270 °C at a rate of 20 °C min^−1^. The injector temperature was set at 270 °C and the detector temperature at 300 °C with N~2~ as carrier gas.
3.2. Metathesis Experiments
---------------------------
All the reactions were performed under a N~2~ atmosphere in a glass reactor fitted with a thermometer and a rubber septum. For the reaction in RTILs, 0.5 mL of the substrate was added to 1 mL of ionic liquid and stirred for 10 min to attain the reaction temperature. An internal standard (0.05 g) was added followed by the addition of 12.4 mg of catalyst. All the catalysts were soluble in ILs with the substrate forming a biphasic mixture with ILs. Samples were withdrawn by a syringe at regular time intervals for up to 4 hours. The reaction was terminated by immediately quenching with a few drops of ethyl vinyl ether \[[@b22-ijms-12-03989]\]. The quenched sample was diluted with solvent and analyzed by GC.
4. Conclusions
==============
For the self-metathesis of methyl oleate, Hoveyda-Grubbs catalyst **4** provided the best result with enhanced activity. Similar conversions were observed in all the three \[bmim\]\[*X*\] type ionic liquids selected. However, when compared with \[bdmim\]\[*X*\] type ionic liquids, higher activity was observed, which shows the superiority of \[bdmim\]\[*X*\] type ionic liquids as reaction media. Furthermore, it was found that Hoveyda-Grubbs catalyst **3** could be reused for three consecutive runs without loss of activity. Overall, the use of ionic liquids in the metathesis reaction opens a route for the use of vegetable oil as a renewable source of raw materials for the chemical industry.
{#f1-ijms-12-03989}
{#f2-ijms-12-03989}
{#f3-ijms-12-03989}
{#f4-ijms-12-03989}
{#f5-ijms-12-03989}
{#f6-ijms-12-03989}
######
Activity of **3** and **4** for the self-metathesis of methyl oleate in \[bmim\] and \[bdmim\]\[*X*\] type ionic liquids.
RTIL Catalyst Temperature (°C) Conversion (%) Selectivity [a](#tfn2-ijms-12-03989){ref-type="table-fn"} (%)
------ ---------- ------------------ ---------------- ---------------------------------------------------------------
5a 1 40 57 99
2 65 97
3 55 99
4 78 96
1 60 59 99
2 75 94
3 57 98
4 87 93
5b 1 40 58 99
2 67 96
3 56 99
4 79 96
1 60 61 99
2 78 94
3 59 97
4 87 95
5c 1 40 57 99
2 64 96
3 54 99
4 74 97
1 60 60 99
2 73 96
3 59 99
4 85 95
6a 3 40 57 99
4 81 98
3 60 60 97
4 90 95
6b 3 40 58 99
4 85 99
3 60 60 97
4 40 92 95
0.5 mL of MO (MO/Ru ratio = 100), 1 mL of IL, reaction time = 4 h;
selectivity towards PMPs.
######
Activity of **1** and **2** for the self-metathesis of methyl ricinoleate.
RTIL Catalyst Temperature (°C) Conversion (%) Selectivity [a](#tfn4-ijms-12-03989){ref-type="table-fn"} (%)
-------- ---------- ------------------ ---------------- ---------------------------------------------------------------
**5a** **1** 60 45 99
**2** 58 99
**5b** **1** 60 41 99
**2** 56 99
**5c** **1** 60 38 99
**2** 60 55 99
0.2 mL of MR (MR/Ru ratio = 100), 0.5 mL of IL, reaction time = 4 h;
selectivity towards PMPs.
######
Recycling of catalysts in \[bmim\]\[BF~4~\] [a](#tfn5-ijms-12-03989){ref-type="table-fn"} and \[bdmim\]\[BF~4~\] [a](#tfn5-ijms-12-03989){ref-type="table-fn"},[b](#tfn6-ijms-12-03989){ref-type="table-fn"}.
Catalyst Conversion (%) Selectivity (%)
---------- ------------------ -----------------
**1** 1st run: 57 (58) 99
2nd run: 56 (56) 99
3rd run: 42 (45) 98
**2** 1st run: 67 (72) 99
2nd run:50 (60) 97
3rd run:45 (57) 97
**3** 1st run: 56 (58) 99
2nd run: 56 (58) 99
3rd run: 49 (53) 99
4th run: 22 (30) 99
**4** 1st run: 74 (85) 99
2nd run: 40 (54) 95
3rd run: 15 (23) 95
All reactions were performed at 40 °C for 4 hours, substrate/Ru = 100;
values in bracket are for \[bdmim\]\[BF~4~\].
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Apple's dominance in the mobile industry became less pronounced in 2012, according to tracking data compiled by mobile ad network Millennial Media. Millennial notes, for instance, that while the iPhone continued to generate the most ad impressions amongst phones, growing 72 percent year-over-year to a 15.59 percent share, Samsung's Galaxy S series grew 182 percent to hit a 4.24 percent share, pushing BlackBerry Curve phones down to third place at 3.91 percent.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Cement slurries used in drilling – types, properties, application
The cement slurry is formed by mixing water with Portland cement or with a mixture of cement containingf various additives. The most important goal of cementing boreholes is to counter the inflow of water into productive levels of rocks with oil and gas. Physico-chemical conditions in borehole have a major effect on the properties of the grouts (bonding, hardening and durability). The most important of these are increased temperature and pressure as well as the presence of reservoir waters, drilling mud and gas migration. The article specifies the requirements for cement grouts used in drilling. The formation process and various types of cement have been characterized due to their phase composition and chemical and mineralogical composition. The treatment methods for the cements of drilling cements are also given. Finally, examples of drilling in the Polish Lowlands are illustrated as a sample of many factors affecting the type, properties and application of cement.
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Invalid Investigation–Derides & Mocks The Death Of An Innocent Man…
An innocent man was killed while in the custody of the Key West Police Department. This disabled 61 year-old met his fate at the hands of police officers, as he laid down on the ground before them. This defenseless and helpless citizen appeared compliant to all commands that were directed at him.
An onslaught of aggression was executed upon Mr. Eimers. He was face down in a prone position, with his arms extended above his head. He did not pose a threat to anyone.
The violent force exerted on him to effect an arrest, was not commensurate with the submissive actions demonstrated by the decedent.
Placing their lives on the line every day to protect and serve our community, law-enforcement is a noble profession. Split-second, life or death decisions are required of them each day. When needed, we value and applaud those who have chosen such a career. On the fateful day in question, their state of mind remains a conundrum.
Unfortunately, I’ve witnessed police misconduct on more than one occasion, which resulted in the loss of life and a subsequent cover-up, to conceal crimes. These heavily armed organizations are capable of malfeasant and murderous actions.
It appears that the conduct of some police officers arresting Mr. Eimers were the source of his death. In one sense, this obedient and compliant individual may have had the life beaten out of him.
Credible evidence indicates that immediately following the death of this elderly man, as he laid lifeless at the feet of his executioners, some police officers allegedly misinformed the responding EMT’s, as to the factual details that caused the subject’s fatal injuries.
Misrepresenting these details and the manner in which Mr. Eimers acquired his injuries, to the arriving paramedics may have cost him his life. Such behavior is a serious breach of the public trust and may constitute a crime.
An independent investigation revealed that rather than gathering witnesses and collecting statements capable of establishing a basis of fact, which described the tragedy that had just occurred, it appears that the Key West Police Department did their best to chase and scatter away all observers to the their actions.
In addition, apparently there was a cadre of vacationing New York City police officers having something to eat, as this incident unfolded before them. It’s suggested that argumentative words were exchanged between this group and the KWPD. It appears that this visiting group vociferously voiced their disapproval of the tactics employed by the arresting officers. It’s my understanding the FDLE and the KWPD did not take any statements, nor record any data attesting to the identity of any witnesses to the event that had just killed an innocent man in their custody.
A subsequent independent investigation suggests that two KWPD officers may have returned to the scene of this fatality, in an effort to silence all employees who witnessed this tragic incident. A witness to this fear tactic, stated that one of the individuals appeared to pose as an FDLE agent. As chronicled by some other witnesses, these alleged investigators utilized veiled threats to invoke the censorship they desired.
It’s apparent from the start that the relationship formed between the KWPD and the FDLE, based upon their own guidelines, protocols, policies and common sense; violated the basic tenets required of any credible and reliable examination of the facts.
According to the findings of an independent examination, Carol Frederick (FDLE Supervisor) of the Eimers’ killing, was contacted by an eyewitness who works for the U.S. District Court system in Maryland via voicemail on January 30, 2014. This woman possessed a plethora of relevant information related to the immediate events surrounding the police action that caused an in-custody fatality.
Evidence suggests that this voicemail was ignored by Ms. Frederick. She apparently refused to review or record this pertinent testimony. This primary sourced evidence was not made a part of the investigation . Along with all the witness statements that were not gathered, this critical evidence was not available to the medical examiner. Apparently, this information will not be included in the findings presented to the ‘grand jury’.
FDLE violated three policies contained within their own Investigative Code of Conduct. Kathy Smith (FDLE ), the ex-wife and mother of Capt. Smith’s child, violated all investigative doctrines when she assumed the duties to examine Mr. Eimers’ death. Her former husband (Capt. Smith) is a Commanding Officer in the KWPD. All of the police officers under scrutiny in the aforementioned killing are linked to Capt. Smith.
As it appears, the questionable conduct and poor judgment exhibited from the start, and throughout the FDLE investigation, constitutes a brazen disregard of jurisprudence. The apparent negligence and neglect in this matter, invalidates their investigation. As such, it cannot be proffered before a ‘grand jury’. To fraudulently proceed with a document that maligns the integrity of the court, would constitute a serious crime.
Reviewing the data secured thus far, it suggests that either the KWPD and the FDLE are attempting to intentionally mislead or they have been grossly negligent.
State Attorney Catherine Vogel, if you do not stop this deception dead in its tracks and immediately initiate an independent investigation, the rats will abandon ship as you take her down to the bottom. This is not solely a Key West issue. Renowned individuals and reputable institutions world-wide are waiting to call in the cavalry, if there are any appearances of impropriety. I believe that you’re underestimating the ubiquitous nature of the depraved indifference, projected by this tragedy.
The Mayor, City Commissioners and Chief Lee have a problem on their hands. A pattern of behavior that appears to be dangerous and reckless (actions that have injured or killed) has been exhibited by their police department. These public officials have not shown any interest in taking a detailed examination of the events that resulted in the Eimers’ killing. This innocent man’s life was taken for ‘no just cause’.
It’s unfortunate that so many individuals appear willing to destroy their careers, reputations and finances, because they are unable or unwilling to professionally adhere to their oaths and codes of conduct. They prefer to roll the dice, rather than lead through the force and might of the character.
A ‘cover-up’ runs long, hard and deep. It exacts a heavy toll on all involved, never relinquishing its maddening grasp on all participants. Relentlessly squeezing the very essence of life from one’s existence.
Denying an American their right to live is a serious crime. Federal and State Prisons host such violators. As the plot thickens, the web of culpability will increase. As the noose tightens around the necks of some, they will squeal.
Madam State Attorney, please head off the discernible harm that will be resultant by your inaction. Enduring the toxicity of recurrent themes and speculation originating from a below par investigation, will serve to satisfy no one, while disrupting all.
Initiating a proactive stance in light of the facts that have been presented, is the right thing to do. Catherine, please do not squander this opportunity to rise and shine for the sake of your community. Assuming a leadership role now, will birth a reconciliation and healing that will be forever remembered and appreciated.
~~~~~~~~~~~~~~~~~
John Donnelly
John Donnelly, a resident of Key Largo was born, raised and went to high school in the South Bronx. Upon graduation he was awarded several scholarships to college. He chose to enlist in the United Sates Marine Corps. While serving in Vietnam John was wounded in action. He received two meritorious promotions, one during combat. Upon discharge and return to America, John had a difficult time transitioning back into civilian life. He found himself homeless for the next 4 years. As he worked out some troublesome concerns, he began to yearn to make some sense of his experiences via education. He sought and received his GI Bill benefits. He applied and was admitted to New York University. He later transfered to the University of Miami where he graduated on the President’s Honor Roll. John secured a teaching position at a Maximum Security Prison Facility for criminally insane adolescents. While working there he earned a Master of Science degree from Florida International University. He graduated Summa Cum Laude. More…
21 Responses to “Invalid Investigation–Derides & Mocks The Death Of An Innocent Man…”
It’s become apparent by now that FDLE exists solely for the purpose of covering for any wrongdoing by police departments in this state. The entire Eimers investigation or lack thereof is the status quo for FDLE. It really pisses me off if taxpayer money funds FDLE because they definitely do not serve the public in this state. If there is not an investigation into this entire sordid mess we need to step up and demand one. We should have the balls to stand up to the KWPD and FDLE to hold peaceful demonstrations until something is done. It’s apparent our worthless Mayor and Commissioners aren’t going to do anything. The only way things will change is if we care enough to do something.
Catherine Vogel MUST step up to the plate. She is the only one left at this level who can find out the truth and let the chips fall from the top down. Citizens should not have to fear their supposed protectors.
The two officers who revisited the employees to stifle them should be charged and fired. Thank you John Donnelly.
Please Catherine. This why I voted and campaigned for you.
Just thought everyone should know something I experienced this morning. I thought I was placing a post on the “Craig Cates For Mayor” Facebook site. My post consisted of my disappointment with our Mayor and Commissioners on their inaction on the Charles Eimers murder at the hands of KWPD. Not only was my post not shown after finishing but I am now not able to make any posts to this site. Obviously “his highness” will only allow back-slapping posts from his “bubba system” friends saying what a great guy he is. Is this really someone you want as Mayor of this island?
I value and appreciate the comments made by ‘Posters’. Thank you very much. [email protected], I sincerely appreciate your kind words…..
Addressing DESERTCOGO’s post, it’s dangerous and disturbing that the Mayor has used the power of his office to censor and punish you for exercising your 1st Amendment Rights.
Petitioning the government to redress the harm, injury and loss of life resultant from the actions of their agents, must never be ignored, silenced or politicized.
DESERTCOGO, please maintain a journal that keeps a record of all transactions with the Mayor’s Office. Retain copies of all communiqués and documents when dealing with them.
If the trajectory of the Eimers’ proceedings are not dramatically altered, they will unravel. State Attorney Catherine Vogel has a duty to intervene. Fellow State Attorneys throughout Florida have not hesitated to conduct their own investigations when faced with similar circumstances.
The citizens of the 16th Judicial Circuit need a State Attorney who will immediately assume a leadership role and lend some clarity, credibility and integrity to the investigative process. If this is not done, Ms. Vogel’s inaction will leave an indelible ‘black mark’ on the City of Key West.
As individuals and outside agencies get their hands on all of the information surrounding the Eimers’ killing, those who were involved in stonewalling and whitewashing this tragedy, will be severely punished. Their consequences will cause them to suffer.
Please keep your records intact. They will be valuable, as those having any culpability in a ‘cover- up’ will be exposed and held accountable….. An analysis is engaged even now…..
bless you john for a very succinct report and time line in the ongoing emiers killing….and killing was what it was from the very start on release of the video 2 weeks after the original police reports. thier conspiracy has grown thick with time. i do hope that ms vogel will do the proper thing and institute a full investigation from soup to nuts and may the chips fall where they may and especially stand her ground…of this i have no comment of speculation since i have no personal understanding of her integrety and ethics. we as interested individuals shall see in due course.
in closing let us not forget the kw citizen newspaper when they finally after many weeks ran the video from about the 20 second mark or last half making sure the first damning part was unseen by the viewer. and let us not forget thier discontinuance of the online story comments function after the overwhelming harbor dredging issue vote and now today a comment below which summarizes my view sent to the citizen yesterday in rebut just to have it edited to the first 2 sentences by thier crack staff of establishment shade pullers! i will rest my case till next time….
“I beg to differ..the Eimers’ case did not damage the reputation of KWPD the cops did a fine job of that all by themselves in damaging thier own reputation! Just look at the video pile on evidence. As for ‘this force is dedicated to keeping our city in check’ since when do ‘we the people’ need to be ‘kept in check’ like some nazi gulag! The cops murdered Charles Eimers with thier jackboot grunt knuckle dragging tactics plain simple and to the point.”
KWPD doesn’t investigate anything…..they just make things go away. And, while I may be biased, the taxi compaies there and thier drivers aren’t much better!! My son died, the taxi driver let him, the police don’t care (just another kid…..) and to be honest niether does any of the media in KW. He was a citizen, he had lived there since he was 17…he moved many, many people there. We were all citiizens (and some still are) at one point but he was just a kid who drank too much…who cares that the taxi driver let him drown in FRONT OF THE POLICE STATION!?!?!? He doesn’t matter…niether does this tourist. Who Cares?!!? “We will just sweep it under the rug and it won’t matter to the outside world”…..that is how I feel KWPD feels about most things!!
I would like to know who came up with “one human family” because the irony of it is just plain laughable. Those in charge of this island better wake up because the Eimers killing has pushed many of us over the edge and we will not put up the crappy way things are done here by crappy politicians. The bubba system that exists on this island is worse than any place I’ve lived and that’s saying something. Mayor Cates censuring me from his Facebook site was the last straw for me. I refuse to support any bubba.
I cannot begin to understand the pain and suffering that you are required to live with on a day to day basis. My deepest and sincerest condolences to you.
The tragedy you described epitomizes a cold and calculating attitude held by some at the KWPD. Individuals such as these have infiltrated every walk of life (taxi drivers/companies, Mayor, City Government, FDLE Investigators and State Attorneys).
The rights and constitutional protections guaranteed to your son appear to have been abandoned. As Mr. Eimers, your son was an American citizen entitled to safe and humane treatment at all times.
I’m extremely honored that you chose to share your painful experience with our readers. Expressing what you are having to endure, provides us with insight into the minds of those who would harm us.
As of yet, I’m unfamiliar with your son’s tragic circumstances. I will attempt to locate a detailed description of this heartbreaking loss.
As tragically reported in a previous post, the absence of character, integrity, competency and courage in our schools, governments and police departments; have clear and convincing consequences.
Catherine Vogel appears to have been doing a good job as our SA. However, she hasn’t really had to face anything of much importance. Her integrity and trustworthiness have not been challenged.
The ‘Eimer’s Case’ has given her the opportunity to step out into the spotlight, assume command and guide her constituents through troubled waters. Thus far, she has failed miserably.
Yesterday, I spoke with several individuals holding prestigious positions at universities in the field of law. Their jurisprudence has distinguished them, when addressing circumstances similar to those facing Ms. Vogel.
It’s inconceivable that Ms. Vogel would pursue a course of inaction. It opens the door for all manner of affliction.
If Ms. Vogel would initiate an investigation, as many of her colleagues and predecessors have done when facing similar contingencies, she would rally the forces of good behind her and galvanize her constituency. It would place her in a position to be revered and honored.
In the time that I’ve lived here I’ve come to the conclusion that none of the bubbas who run this island can be trusted. I’ve also come to the conclusion that no one, including Ms. Vogel wants to rock the boat for fear of retribution from the bubbas. There is absolutely no reason why there has been no final word from FDLE regarding the murder of Charles Eimers by KWPD. I’m sure they have already decided there was no wrongdoing by KWPD because that is what they do, cover for law enforcement. I sense they are dragging their feet with the results of their investigation because of local unrest here. We all owe a debt of gratitude to the Blue Paper for not allowing this tragedy to be forgotten like all of the rest of Gestapo tactics of our local police department.
A woman with a Spanish name sent me something John Donnelly had published in the blue paper about the Eimvers death FDLE investigation, and I sent her the blue paper link to the longer video and asked if she could translate it for me? She replied this morning, and I included that in today’s post at goodmorningkeywest.com.
Sloan,
It sounds like the woman in the video is saying: ” Oh God, look at all the police, look at all the patrol units; they won’t let him….”. It ends right there.
With thanks,
Alicia
Naja replied to me:
“Arnaud couldn’t make out what the woman who shot the video was saying – other than oh my god and exclaiming about all the police. The line you put in bold he struggled over – he just wasn’t sure what she was saying. Interesting that your translator says she said ‘they won’t let him’. I find the woman’s voice to be a distraction – I purposefully removed it from the original video posted to youtube. I gave the video to the attorneys, however, with audio intact. She wasn’t close enough at the point where she’s shooting the footage to offer more information than the video itself shows us – so I really don’t place a lot of importance on her commentary Found in the video. I have tried to no avail to contact her to know if she has any more information but the person who gave it to us said she had no contact info and has stopped communicating with me. I believe that person was interviewed by FDLE. Wish the Spanish speaking woman would come forward and there were witnesses in the restaurant who had a much better view and some of them even live here. Wish they would all come forward too. Naja”
I wondered how Naja could decide the Spanish woman’s voice was a diversion, when Naja did not know what the Spanish woman was saying? Then, I got a bad sinking feeling.
Metaphysically, Naja inadvertently had sent out a message into the spirit realms that the contemporaneous remarks of an eye witness to what happened on South Beach last Thanksgiving Day were a diversion. In spirit, every eye-witness heard that message and were inadvertently encouraged not not to come forward, eye-witnesses such as the NY City police officers, the Spanish woman, and the person who shot the video.
Later, I got another bad sinking feeling. The message inadvertently encouraged the the FDLE investigators and Cathy Vogel not to try to locate and interview all eye witnesses. And the message inadvertently encouraged the KWPD officers involved, Police Chief Donie Lee, and FDLE to alter evidence and/or cover up what really happened.
I was sick to my stomach, because I knew that’s how the interface between people and the spirit realms works. Call it karma, if you wish. Naja altered the first video which she shared with the public; how the investigation went after that was the natural metaphysical consequence.
I published this perspective twice at http://www.goodmorningkeywest.com, but that was not enough to break up what is blocked in the spirit, because this perspective needs to be published in the blue paper where the blockage originated. So far, Naja has declined to publish this perspective, which I twice have submitted in different forms. I hope she will publish it now.
No Sloan. What is preventing the eye-witnesses from coming forward and what is/has prevented Vogel from performing a legitimate investigation is their lack of conscience. They have shrunk away from their human responsibility just as you try to shrink away from yours by claiming that your angels are in charge of everything.
That, too, Sister. But the metaphysics also is in play. I know how far out seems, but it isn’t nearly as far out as some things I have personally seen happen, sometimes to me, sometimes to other people. I knew people before I was moved to Key West by the angels who run me, who would have agreed with what I submitted. I know a woman in another state right now, who dreams about the Eimers case,and other things I’m engaging down here, who sees what I described. It’s not strange to her. What I wrote, Sister, describes work I used to do a lot of before I was moved to Key West. Sometimes that work comes back around in stuff I handle down here. Sometimes I write about that kind of work at goodmorningkeywest.com, sometimes I don’t. The human and the spirit realms are interlaced – one human family seriously expanded. Everything is connected. Whatever happens in human ways affects the spirit realms, and whatever happens in spirit ways affects the human realms. Imagine a big spider web. You pluck a strand of the web. The vibration is felt in and affects every strand of the web.
They are your angels, too, Sister. They ran the Eimers case to Naja and Arnaud. They have run other things to Naja and Arnaud, and will run other things in the future. They provided the bystander with the cell phone and the video. They kept Eimer’s body from being cremated. They arranged for Officer Lovett to boast to a friend what he had done to Eimers, not knowing the friend would rat him out. They brought a very reluctant lawyer into the case, David Horan told me himself that he didn’t want to take the case; he was a friend of Lovett’s parents, had known him since he was a little boy, was close to him, but he felt like he had to take the case anyway. Horan told me he was reluctant for a long time to accept the reason Eimers died was because the cops thought he was homeless. That was a real stretch, Horan told me, for him to come to that view. The angels brought Horan to that view. He may never believe that. You may never believe it, Sister. But I have seen the angels work lots of situations in which I was involved. I’ve seen lots of strange things happen. Either you can see such things, or you can’t, has been my experience. I have known quite a few people who saw such things. It’s a dance of the human realm with the spirit realm. The two cannot be separated, but many people live as if the two are separated, or as if the spirit realm does not even exist.
Allrighty then Sloan. I sure hope those angels decide to make some arrests for the murder of Charles Eimers and for the felonies involved in trying to cover up the murder. Cause it sure doesn’t look like any mortal beings are gonna do it.
Roger that, Sister. I told the angels they stuck me way out on a skinny limb in a high tree, and I now it’s on them to get the “prosecution” moving. However, I would not dare predict what they might or might not do. They play their cards real close to the vest until they are ready to turn them face up, has been my experience with them.
Privately, John Donnelly kept encouraging me to keep submitting my thoughts. I wrote to him last night and told him my third submission had been cleared from moderation and was showing in reader comments under his article. On waking this morning, I found this email from John in my inbox:
I read in the Citizen that the city attorney says that police officers have a right to “qualified immunity” that protects them from liability while doing their jobs. I believe they lost that right when they flat out lied about Charles Eimers running and collapsing after exiting his vehicle. That was before the video surfaced showing that never happened. What about the officers dispersing of witnesses and basically threatening employees at the restaurant? If they were just “doing their job” why the need to lie and cover-up what they did?
| null |
minipile
|
NaturalLanguage
|
mit
| null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.