text
stringlengths 0
7.84M
| meta
dict |
|---|---|
Ingleside Road station
Ingleside Road is a Tide Light Rail station in Norfolk, Virginia. It opened in August 2011 and is situated along Ingleside Road. The station primarily serves residents of adjacent neighborhoods.
References
External links
Ingleside Road station
Category:Tide Light Rail stations
Category:Railway stations in the United States opened in 2011
Category:2011 establishments in Virginia
|
{
"pile_set_name": "Wikipedia (en)"
}
|
996 So.2d 1084 (2008)
STATE ex rel. David GARRETT
v.
STATE of Louisiana.
No. 2008-KH-0485.
Supreme Court of Louisiana.
November 14, 2008.
Denied. La.C.Cr.P. art. 930.8; State ex rel. Glover v. State, 93-2330 (La.9/5/95), 660 So.2d 1189; State v. Parker, 98-0256 (La.5/8/98), 711 So.2d 694.
|
{
"pile_set_name": "FreeLaw"
}
|
Much of the renewed focus on energy conservation and going green has been on what to do at home and in the workplace. But how about the places where people go to relax and play—resorts and other vacations spots? Are they becoming more green? If, in the last year, you were one of the more than 38 million people who stayed an average of five days in one of the 124,270 hotel rooms and basked in the glow of 15,000 miles of neon lights and/or nearly 200,000 slot machines, then you have been in Las Vegas and contributed to the doubling of the rapidly growing local population. Las Vegas is ranked as the second sunniest city in the United States with sunshine 85 percent of the time and a maximum monthly temperature above 80°F for half the year.
That should make it no surprise, though the state population is ranked 35th from the largest, Nevada had the fastest increase rate of electricity consumption of any state in the country (4.5 percent per year) until 2009. According to statistics collected by the U.S. Department of Energy (DOE), it had a similar top growth rate percentage in population as well. With more visitors to the state’s two resort cities known for their casinos, it seems reasonable that much of the more than 35 billion kilowatt-hours of electricity annually consumed runs these operations. Evidence of such comes from Sue Roaf, a professor and expert in environmental architecture, who determined that residents and guests at the new City Centre near the MGM Grand are projected to consume twice the national household average. TheFunTimesGuide.com reports this would result in a combined annual output of 160 million tons of CO2. Despite some local sentiment that such figures are misleading because most of the electricity comes from the hydroelectric generators of the Hoover Dam, only 17 percent of the state’s electricity is hydroelectric.
The intent is certainly not to single out Las Vegas. But it does serve as a good example for the old adage, “When you have lemons, make lemonade.” To start with, the abundant sunshine that contributes to the high temperatures makes the area an excellent candidate for photovoltaic electricity generation. Unfortunately, the profile of the buildings, with low roof-to-total-volume ratios makes roof-mounted panels inadequate for a significant portion of the energy needed. However, one doesn’t have to drive far from the Strip to find open desert with unobstructed sunshine. A cooperative-based solar generating plant could share the cost and help green the desert.
Considering that the majority of the electricity is being used for cooling and lighting, a large-scale joint venture to convert to more efficient lighting, such as compact-fluorescent and LED-based, and high-efficiency heating, ventilating and air conditioning (HVAC) systems could reduce the premium costs that currently result from such technologies. Whereas the casino floors operate 24/7, sectionalizing the areas to meet the patron loading would allow for reduced lighting and cooling requirements in the off-peak times.
Many people on vacation forget about turning lights off and setting back the thermostat when the room is not occupied. The use of occupancy-sensor-based switches (common in Europe) could be extended from controlling lighting to controlling the cooling/heating as well. Or, thermostats with an “I will be out for X hours” button would allow energy--conscious visitors to help in the cause. Simply closing the curtains rather than letting the sunshine pour into an empty room will significantly reduce the HVAC load. If the occupants don’t cooperate, it could become a daily part of the housekeeping tasks. In many hotels, housekeepers currently open up the curtains or blinds when they clean the rooms.
Other resort areas have different opportunities. According to CityTownInfo.com, half of the Top 10 cities and towns with the highest hotel density are in ski country. While the electrical loads there are more for snow making and chair lifts than glimmering lights, there are opportunities in these applications for higher efficiency motors coupled with adjustable speed drives. As with the energy savings in the desert-based hotel rooms, similar concepts are applicable, though more associated with heating than cooling. Turning down the heat and closing curtains to keep heat in and cold out works just as well when not occupying the room, along with the use of energy-efficient lighting in the rooms and on the slopes. Some of the hotels in the ski resorts also command a much higher daily rate ($400–$1,455) than in the casino hotels. Clearly, passing on the additional cost for energy-efficient heating and lighting improvements would have little impact on the total vacation costs.
Plenty of ways exist to be green while at play. Hopefully, it will soon be a way of life for us no matter where we are.
BINGHAM, a contributing editor for power quality, can be reached at 732.287.3680.
|
{
"pile_set_name": "Pile-CC"
}
|
Q:
EditText clean on first touch issue
I have an EditText that contains a grey text that indicates what the user have to input, when clicked the first time this text must be cleared to allow the user to input the content without waste time to clear the hint text.
so I have added a listener with this code:
if(!editCountryEdited){
editCountryEdited=true;
editCountry.setText("");
//...
}
But unfortunately works only is the EditText is clicked a second time after the first click. How could I fix this?
A:
Use hint http://developer.android.com/reference/android/widget/TextView.html#attr_android:hint
<EditText
...
android:hint="Your grey text here" />
|
{
"pile_set_name": "StackExchange"
}
|
Q:
mysql insert record before last
I have a small problem
I have a database with the following records
tag
---
a
b
c
eof
eof must always be the last record
how can I insert a field before
EOF to avoid this situation?
if i delete eof and insert a field and
after insert eof, i found this strange
situation, mysql writes eof before ??
tag
---
a
b
c
eof
d
I tried it with order by but does not work
I read the record positionally
thanks in advance
A:
This will return eof always last
select * from tsynapse
order by case when tag = 'eof'
then 2
else 1
end asc, tag asc
|
{
"pile_set_name": "StackExchange"
}
|
Citation Nr: 1331553
Decision Date: 09/30/13 Archive Date: 10/02/13
DOCKET NO. 10-15 186 ) DATE
)
)
On appeal from the
Department of Veterans Affairs Regional Office in Roanoke, Virginia
THE ISSUE
Entitlement to service connection for fibromyalgia.
ATTORNEY FOR THE BOARD
Evan M. Deichert, Counsel
INTRODUCTION
The Veteran served on active duty from May 1979 to October 1981.
This matter is before the Board of Veterans' Appeals (Board) on appeal of a February 2009 rating decision of the Huntington, West Virginia, Department of Veterans Affairs (VA) Regional Office (RO) which denied the Veteran's claim for entitlement to service connection for fibromyalgia.
On her March 2010 Substantive Appeal, the Veteran indicated that she wished to present testimony before a member of the Board. In June 2012, the Veteran wrote that she no longer desired a hearing; her request is considered withdrawn.
In March 2013, the Board remanded the case for further development, which has been completed. Stegall v. West, 11 Vet. App. 268, 271 (1998).
FINDING OF FACT
The Veteran did not suffer from fibromyalgia during her active service or for many years thereafter, and there is no competent evidence linking her fibromyalgia to her active service or to any in-service event.
CONCLUSION OF LAW
The criteria for service connection for fibromyalgia have not been met. 38 U.S.C.A. §§ 1101, 1112, 1113, 1131, 1154, 5103, 5103A, 5107(b) (West 2002); 38 C.F.R. §§ 3.102, 3.159, 3.303 (2013).
REASONS AND BASES FOR FINDING AND CONCLUSION
I. Duties to Notify and to Assist
Upon receipt of a complete or substantially complete application, VA has a duty to notify and assist claimants in substantiating a claim for VA benefits. VA must notify the claimant of the information and evidence not of record that is necessary to substantiate a claim, which information and evidence VA will obtain, and which information and evidence the claimant is expected to provide.
An October 2008 letter satisfied the duty to notify provisions. 38 U.S.C.A. § 5103(a); Quartuccio v. Principi, 16 Vet. App. 183, 187 (2002); 38 C.F.R. § 3.159(b)(1). This letter also notified the Veteran of regulations pertinent to the establishment of an effective date and of the disability rating. Dingess/Hartman v. Nicholson, 19 Vet. App. 473 (2006). The claim was subsequently readjudicated, most recently in a June 2013 supplemental statement of the case. Mayfield v. Nicholson, 444 F.3d 1328 (Fed. Cir. 2006).
In any event, the Veteran has neither alleged nor demonstrated any prejudice with regard to the content or timing of the notices. See Shinseki v. Sanders, 556 U.S. 396 (2009); see also Mayfield v. Nicholson, 444 F.3d 1328, 1333-34 (Fed. Cir. 2006).
The Veteran's service treatment records, VA medical treatment records, and private treatment records have been obtained. 38 U.S.C.A. § 5103A, 38 C.F.R. § 3.159. The Veteran has not indicated, and the record does not contain evidence, that she is in receipt of disability benefits from the Social Security Administration. 38 C.F.R. § 3.159(c)(2). A VA examination was conducted in October 2012; in its March 2013 remand, the Board found the opinion in this examination to be inadequate.
The Board remanded the Veteran's claim for a new examination. The Veteran did not attend her scheduled examination and has not provided good cause as to why the examination should be rescheduled. Absent good cause, the Veteran's claim must be decided based on the evidence of record.
There is no indication in the record that any additional evidence, relevant to the issue decided, is available and not part of the claim file. See Pelegrini v. Principi, 18 Vet. App. 112 (2004). As there is no indication that any failure on the part of VA to provide additional notice or assistance reasonably affects the outcome of the case, the Board finds that any such failure is harmless. See Mayfield v. Nicholson, 20 Vet. App. 537 (2006); see also Dingess/Hartman, 19 Vet. App. at 486; Shinseki v. Sanders/Simmons, 556 U.S. 129 (2009).
II. Entitlement to Service Connection for Fibromyalgia
Service connection will be granted if it is shown that the Veteran suffers from a disability resulting from personal injury suffered or disease contracted in the line of duty, or for aggravation of a preexisting injury suffered or disease contracted in the line of duty, during active military service. 38 U.S.C.A. § 1131; 38 C.F.R. § 3.303. Disorders diagnosed after discharge will still be service connected if all the evidence, including that pertinent to service, establishes that the disease was incurred in service. 38 C.F.R. § 3.303(d) ; see also Combee v. Brown, 34 F.3d 1039, 1043 (Fed. Cir. 1994).
To establish service connection, there must be a competent diagnosis of a current disability; medical or, in certain cases, lay evidence of in-service occurrence or aggravation of a disease or injury; and competent evidence of a nexus between an in-service injury or disease and the current disability. Hickson v. West, 12 Vet. App. 247, 252 (1999); see Jandreau v. Nicholson, 492 F.3d 1372 (Fed. Cir. 2007).
Competent medical evidence is evidence provided by a person who is qualified through education, training, or experience to offer medical diagnoses, statements, or opinions. Competent medical evidence may also include statements conveying sound medical principles found in medical treatises and statements contained in authoritative writings, such as medical and scientific articles and research reports or analyses. 38 C.F.R. § 3.159(a)(1).
Competent lay evidence is any evidence not requiring that the proponent have specialized education, training, or experience. Lay evidence is competent if it is provided by a person who has knowledge of facts or circumstances and conveys matters that can be observed and described by a lay person. 38 C.F.R. § 3.159(a)(2). This may include some medical matters, such as describing symptoms or relating a contemporaneous medical diagnosis. Jandreau, 492 F.3d at 1377. When considering whether lay evidence is competent the Board must determine, on a case by case basis, whether the Veteran's particular disability is the type of disability for which lay evidence may be competent. Kahana v. Shinseki, 24 Vet. App. 428 (2011); see also Jandreau, 492 F.3d at 1376-77.
In determining whether service connection is warranted for a disability, VA is responsible for determining whether the evidence supports the claim or is in relative equipoise, with the veteran prevailing in either event, or whether a preponderance of the evidence is against the claim, in which case the claim is denied. 38 U.S.C.A. § 5107; Gilbert v. Derwinski, 1 Vet. App. 49 (1990). When there is an approximate balance of positive and negative evidence regarding any issue material to the determination, the benefit of the doubt is afforded to the claimant.
The Veteran currently suffers from fibromyalgia. She was first diagnosed as suffering from this condition in March 1993. Private treatment records reflect that she was again diagnosed as suffering from this condition in September 2005, and the examiner from her October 2012 VA examination also diagnosed her as suffering from fibromyalgia.
There is no evidence that the Veteran suffered from fibromyalgia during her active service. Service treatment records contain no evidence of her being diagnosed with, or seeking treatment for, this condition. The Veteran acknowledged in a January 2013 letter that she was not diagnosed as suffering from this condition during her active service.
Despite the lack of corroborating in-service evidence, service connection for fibromyalgia may still be warranted if the evidence establishes that the condition was incurred in or is otherwise related to her active service. 38 C.F.R. § 3.303(d).
In a January 2013 letter in support of her claim, the Veteran contended that her fibromyalgia (like her service-connected degenerative arthritis of the lumbar spine) is related to two in-service falls that she suffered. She argued that, since there is no known cause for fibromyalgia, her condition must be related to the falls she suffered during her active service.
Service treatment records document these falls. A June 1981 record in which the Veteran complained of back pain noted that she fell in October 1979 and experienced pain in her coccygeal area that resolved in two weeks. An April 1980 X-ray report reflects that the Veteran complained of pain in her low back since a fall; the X-ray was reported to be normal.
There is, however, no competent evidence of a nexus between her current condition and her active service.
The Veteran underwent a VA examination in October 2012. After examining the Veteran, the examiner concluded that it is less likely than not that the Veteran's fibromyalgia is related to her active service, as there is no evidence of the Veteran's suffering from this condition during her active service.
In its March 2013 remand, the Board found this opinion to be inadequate and remanded the Veteran's claim in order that she be afforded a new examination. Such an examination was scheduled for April 2013; the Veteran did not report for this scheduled examination.
Pursuant to VA regulation, "when a claimant fails to report for an examination scheduled in conjunction with an original compensation claim, the claim shall be rated based on the evidence of record." 38 C.F.R. § 3.655(b) (2013). The Board may order that the Veteran be rescheduled for an examination if the claimant shows "good cause" for missing the examination.
Here, the Veteran has provided no information for why she missed her examination or why that examination should be rescheduled. As no good cause for missing the examination has been presented, the claim must be decided on the evidence of record.
The evidence of record, beyond the Veteran's assertions, is silent as to an etiology for the Veteran's fibromyalgia. Neither the private nor the VA treatment records contain any information regarding the etiology of the Veteran's fibromyalgia.
The only evidence in support of the Veteran's claim is her own argument that her condition must be related to her active service. The Veteran, however, is not competent to opine as to the etiology of her fibromyalgia. Unlike simple conditions (such as a broken leg), determining the etiology of fibromyalgia requires specialized medical training and is not susceptible to lay observation. Jandreau, 492 F.3d at 1377. Accordingly, the Veteran's contention that her fibromyalgia is related to her in-service falls is not competent evidence of such.
The preponderance of the evidence is against the claim for service connection for fibromyalgia; there is no doubt to be resolved; and service connection for fibromyalgia is not warranted.
ORDER
Entitlement to service connection for fibromyalgia is denied.
____________________________________________
RONALD W. SCHOLZ
Veterans Law Judge, Board of Veterans' Appeals
Department of Veterans Affairs
|
{
"pile_set_name": "FreeLaw"
}
|
If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Western Mountaineering is all you'll ever need. I also checked out the Mont-Bel stuff which can sometimes be had a for good price. Most important, try to test it out. Get in it. I found they are all cut so differently that you really need to try them.
x2...WM makes a great lightweight warm bag. They are a little pricey but when it is 10 degrees out and you are not cold you won't think anything of it.
I love my BA pad sleeve system as I also cannot do the mummy and get a good nites sleep cause I move around too much. In a mummy, it takes me so long to squirm around and get restraighted. I've had the BA about 3 yrs now and although not perfect, it's a major improvement for me!
I picked up a big agnes zirkel a few years back due to the sizing looking nice. I am a bigger guy and like my space. I have always hated being confined to a mummy bag. However, when I put the pad in the slot, it made it too confining for me by holding the entire bottom of the bag flat(I am a side sleeper/roller). So, I would use it just on top of the pad which was okay unless it got below freezing and the cold would come in where the uninsulated back part rode up. Eventually, I sold the bag and have since found Montbell. I have a couple of their UL SS downhuggers in a couple flavors and I love them. They have some stretch to them which gets me passed the confined feeling and I can roll over without cold spots. Check them out if you get a chance.
If you want a Big Agnes, but dont want to pay to much to try one out, look at this website. They have used bags for great prices. I picked up a BA Horsethief for $80 and love the bag.http://www.bwear.com/home.php?cat=364
This is my first time trying the Big Agnes and was skeptical about the pad system. It is very nice. I did have one cold spot that the baffle didn't seem to seal out. I also have numerous bags from Marmot (Couloir, Axiom, DriClime Maverick, and a Kid bag). Also have a Mammut bag. It induces claustraphobia, but is warm for its weight and packs super small. If I had to choose one pack to cover most of my needs I would probably go with the Helium Membrain by Marmot.
I have a Valandre Mirage and love it. It is a 20 degree bag easy. The big three in sleeping bags are Valandre, Western and Feathered Friends. You buy one of those 3 bags and it will be the last one you buy.
|
{
"pile_set_name": "Pile-CC"
}
|
Slideshow ( 5 images )
TOKYO (Reuters) - Japanese telecommunications company NTT is hoping to bring an added dimension to the 2020 Tokyo Olympics by projecting live sporting events with its 3D hologram technology.
NTT has been working on a new system it calls “Kirari! For Arena”, which, unlike its current technology, does not require sensors to be attached to an athlete’s body.
The company is aiming to have the system ready by the time Tokyo hosts the next Summer Games.
“From all four directions, you can see players projected in the field moving in 3D,” Kota Hidaka, senior research engineer at NTT Service Evolution Laboratories, said at a demonstration on Thursday.
“Kirari! For Arena” uses multiple cameras to track the movements of the player. Images are then transported to devices which show 3D hologram figures real-time in a different location.
So, for example, fans unable to watch a karate match in an arena could watch 3D holograms of the fighters battling it out live.
“We are hoping to establish the system and deliver the service in which people can enjoy watching games by gathering around the 3D hologram,” Hidaka said.
NTT is an official sponsor of the 2020 Tokyo Summer Olympics.
|
{
"pile_set_name": "OpenWebText2"
}
|
Sen. Warren is tribune of left and its fondest White House wish
Amanda Knief asked the question with a laugh, but liberal fervor for Sen. Elizabeth Warren runs deep, as evidenced by the fans who’d gathered some 90 minutes before the Massachusetts Democrat took the stage in Detroit recently at the nation’s largest gathering of liberal activists.
Volunteers with a group that’s hoping to convince Warren to launch a bid for the White House handed out “Warren for President” stickers and hats. As the senator signed copies of her latest book, “A Fighting Chance,” supporters pressed her to consider a run in 2016.
“She represents the conscience of America,” said Monica Lewis Patrick, 48, a public policy analyst in Detroit.
She cited Warren’s unabashed criticism of the banking industry, which the senator charges duped homeowners and helped spark the financial crisis.
“She speaks truth to power unapologetically,” Patrick said.
Knief, 37, a public policy activist, said Warren was inspiring a new round of activism.
“She makes us think, ‘We can do this,’ ” said Knief. “She’s out there making things happen.”
A former Sunday school teacher, Warren delivers populist stemwinders with confrontational rhetoric. She speaks more like an activist than a politician, decrying “sleazy lobbyists” and their “Republican friends in Congress,” along with rapacious corporations and big banks that she says “swagger through Washington” focused on squashing any effort to regulate them.
For liberals and many who feel left behind by an economy that’s supercharged Wall Street but done little to lift salaries, the subjects of Warren’s ire are fitting targets.
“Most of our politicians are preoccupied with raising money, much of it from Wall Street,” said Egberto Willies, 52, a Houston blogger and software developer. “We finally have a politician who truly believes and advocates for the middle class.”
Warren comes by her populism honestly. She taught for two decades at Harvard Law School, but the Oklahoma native underscores her modest roots: her father was a janitor and her mother worked at Sears, earning minimum wage, Warren tells audiences.
Her parents didn’t have enough money to send young Elizabeth to college, so she enrolled at a commuter school, paying $50 a semester.
Warren first drew widespread liberal ardor when she issued pointed critiques as the chair of a congressional oversight panel that monitored the $700 billion federal taxpayer bailout of the banking industry in 2008. At one hearing, she allowed protesters toting signs that read “Give us our $$$$$ back” and “Where’s our money?” to stay during a nearly two-hour congressional hearing.
She left the panel in 2010 to help launch the Consumer Financial Protection Bureau, aimed at cracking down on deceptive and abusive loan products. Calling for a strong consumer watchdog, she memorably said her second choice would be “no agency at all and plenty of blood and teeth left on the floor.”
Many had hoped she’d become the bureau’s first director, and she was a natural choice. But President Barack Obama passed her over amid staunch opposition from congressional Republicans and financial service industry lobbyists who thought Warren was hostile to their interests.
She ran for the Senate instead, defeating Republican former Sen. Scott Brown, despite what she says were pledges by Wall Street to “send money by the double bucketful” to defeat her.
Warren boasts that the financial protection agency has returned $4 billion to consumers, and as much as some backers would like her to run for president, many want her to remain in the Senate. She’s repeatedly told reporters pressing her about presidential aspirations that she’s serving out her Senate term, which runs to 2019.
“I am focused on the 2014 race,” she told reporters after a fundraiser for Rep. Gary Peters, D-Mich., who’s running for the Senate “It is absolutely critical to this country. We can’t get distracted from that.”
Warren’s populist appeal has made her one of her party’s most valued surrogates on the political circuit. She’s raised $2.6 million for candidates already and has stumped for Democratic Senate hopefuls, even in red states such as Kentucky and West Virginia.
Whether or not she’s firmly committed to not seeking the presidency in 2016 _ the Democratic contest is frozen, in any case, while it waits for a decision by Hillary Clinton _ Warren’s fundraising and book tour gives her a chance to see how well her populist road show plays around the country.
The Progressive Change Campaign Committee, which bills itself as the “Elizabeth Warren wing of the Democratic Party,” says there’s a growing audience for her message. It plans to use Warren as its standard-bearer and to pressure Democrats who seek the White House in order to see whether they, like Warren, support more Wall Street regulations and expanding Social Security.
“There’s a rising economic populist tide in America, and Elizabeth Warren is the personification of that movement,” said Adam Green, the committee’s co-founder.
Republicans have taken note of Warren’s popularity. America Rising, a conservative group that does opposition research on Democrats, says it’s tracking Warren now, along with Clinton.
“Democrats are launching a campaign to draft liberal Democrat Elizabeth Warren for president,” America Rising said in a fundraising appeal. “America can’t afford to let that happen.”
Republicans have seized on the consumer bureau’s plans to renovate its Washington headquarters for as much as $216 million as a sign of government waste.
Warren enthusiasts, who can cite her stances on issues as easily as movie buffs detail their favorite actors’ starring roles, say critics are aiming at the senator to avoid oversight.
“If we didn’t have her voice in Congress, as one-sided as it is, it would be a disaster,” said Lois McCarthy, 67, a retiree from Maryland’s Eastern Shore. “She says exactly what needs to be said, what everyone else is ignoring.”
McCarthy said she’d heard about Warren’s push for the consumer protection board, followed her Senate race and now tracked her Capitol Hill career via Facebook.
“We bailed out these big banks, but you don’t see them bailing out people,” McCarthy said. “Someone has to be vigilant for the consumer, and that’s Elizabeth Warren.”
Read Next
President Donald Trump proposed deal to end the shutdown by trading border wall funding for protections of over a million people living in the United States. Supporters accused him of abandoning his base.
|
{
"pile_set_name": "Pile-CC"
}
|
Disclosure Policy
October 11, 2009
Loving the Ocean
I've been thinking about the ocean a lot lately. I'm tempted to set up an aquarium in our living room full of saltwater fish and live coral. I see so many beautiful fish every time my husband and I make a run to the pet store. I would love to have a couple Reef tanks filled with some tropical fish. When I was younger, we had a tank with fish and it was so relaxing just to watch them swim. Maybe I liked The Little Mermaid a little too much as a kid because it seems like all my life I've wanted to live in or at least near the water. There is something about swimming that I find peaceful. I'm a lover of living creatures and I'm kind of surprised that I don't have a fish tank. I have been telling my husband that I need some frogs and turtles that swim. I also want a lizard that can run around my house. Several people act a little shocked when I say that I would like to have a lizard. I don't know why some people don't like them. I wasn't allowed to have one a kid because my mom doesn't like them. One day I'll have a whole house full of living creatures.
|
{
"pile_set_name": "Pile-CC"
}
|
Comparative analysis of VP8* sequences from rotaviruses possessing M37-like VP4 recovered from children with and without diarrhoea.
Rotavirus strains belonging to G types 1 to 4 and having a P3 genotype (M37-like VP4) were recovered from children with symptomatic and asymptomatic infections. Partial sequences of their VP4 genes were determined in an attempt to characterize these strains further. The genomic regions encoding VP8*, the connecting and putative fusion peptides and three other regions in VP5* were sequenced. The deduced amino acid sequences were compared with rotavirus strains belonging to different P genotypes that had been previously reported. High degrees of identity were found between the VP8* fragment of all human P3 strains (92.7 to 99.7%) suggesting that they belong to the same genotype, regardless of differences in their virulence. Furthermore, based on comparative sequence analysis, we did not identify any amino acid(s) that differ appreciably between symptomatic and asymptomatic strains and could therefore be associated with virulence. The results suggest that the P3 genotype, although frequently associated with asymptomatic infections, may not be the single determining factor in attenuation of symptoms.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Q:
Why do multiplayer games/chat services ever need to be able to open all NAT ports?
I understand that games need to be able to open NAT ports in order to receive incoming connections.
What I do not understand is why there is an issue between "limited" and "open" NAT. The only reason I can see that a game would ever need to have the flexibility to open any port is if it is intending to accept incoming connections on ports that aren't specified in advance as part of the protocol. (If there was a restricted set specified in advance, forwarding these and no others should satisfy all of the game's demands for NAT forwarding, meaning that "open" would not need to be a thing.)
This seems to imply there's some port negotiation going on as part of the protocol (otherwise how does the connector know that this port, not specified in advance, is available?) but if a connection is already established to conduct the negotiation, why move to a different port? And even then, why can't negotiation be limited to a subrange of ports rather than the entire range?
A:
Just to make sure we're all on the same page in terms of terminology:
What is NAT?
NAT is an acronym for "Network Address Translation". It's a system designed to allow a router to allow a whole private network of arbitrary size to appear to be a single device, from the point of view of the outside world. And the magic is that it does this completely transparently, without the clients on either the private or the public network ever needing to know it's happening.
In effect, it takes outgoing requests from the private network and rewrites them to appear to be coming from its single public address and some public port number, and then when it receives a response, it rewrites the response to forward it back into the private network, to whatever device sent the original request.
This was primarily needed because of the rapidly depleting supply of public IPv4 addresses as the Internet's popularity was taking off, but also brought some moderate security benefits by putting a largely one-way barrier between potentially unsecured consumer devices and the public internet. (It also dramatically slowed adoption of IPv6, but that's a topic for another day)
NAT types
This particular classification of NAT types appears to have originated with Microsoft, where it was identified as useful during development of the original XBox game console, since it classed different NAT types into a useful way to determine which clients would be able to directly communicate with which other clients.
Sony uses essentially the same classification system, but uses different terms. In the discussion that follows, I'll mention both company's terms, and then use them interchangeably. But almost everybody has settled on using Microsoft's more human-readable terms for these NAT types; even Nintendo uses Microsoft's terms, now. The only holdout I'm aware of is Sony, who still use their own terms. (I once spent an enjoyable afternoon with a Sony dev rep, trying to make him slip up and use Microsoft's terms for NAT types. I think he figured out what I was doing, and I never managed it. But I did get him to actually say 'Xbox' once, which got him a stern look from his manager. That was a good day. I am a simple man of simple pleasures.)
"Open" NAT (Microsoft's term) or "Type 1" NAT (Sony's term) describes NAT behaviour where a particular private IP address and port behind the NAT device gets mapped to a single public address and port on the outside network. That is, anybody anywhere on the outside who sends a packet to that public address and port will have it successfully delivered to the inside, no matter what. This means that not having a NAT device gets categorised as "Open" NAT. Alternately, setting up a NAT device to forward the correct ports can also achieve this same result.
"Moderate" NAT (Microsoft's term) or "Type 2" NAT (Sony's term) describes NAT behaviour where a particular private IP address and port behind the NAT device gets mapped to a single public address and port on the outside, exactly the same as in "Open" NAT, except that incoming packets to that public IP address and port will only be delivered if we've already recently sent a packet OUT on that port, to the person who is now trying to send a packet IN. That is, people are allowed to respond to us, but not to initiate communication. (This is the most common sort of NAT). We use hole-punching to get through this; a central server tells both of us the other person's public address and port, and we then spam messages to each other, so both of us pretend that we're initiating the conversation, and our NAT devices let both sets of messages get through, thinking that they're "responses".
"Strict" NAT (Microsoft's term) or "Type 3" NAT (Sony's term) describes NAT behaviour which doesn't meet the requirements of Type 1 or Type 2. That is, a single private address and port sometimes get mapped to different public addresses and/or ports, most commonly based upon who the packets are being sent to. For example, if I send a packet from private port 80/UDP to person A and a packet from that same private port 80/UDP to a different person B, then A and B will see those packets as having come from different public port numbers on the NAT device. This means that A and B can't tell each other how to contact me, because my NAT device is going to be requiring that they each only reply to me using the special public port which it allocated for communication to each of them independently; any packets that A might send to the port which was allocated for talking with B won't get through, and vice-versa.
NAT types are broad generalisations
As mentioned above, there are far more types of NAT than just these three, and many more ways to classify them. Sometimes they're classed as "symmetric", or "conical", or in other ways.
One particularly relevant difference, even inside this classification: Moderate/Type 2 NAT devices often differ in terms of how recently we have to have sent an outgoing packet, in order to accept incoming responses. If there's a long delay between our last packet out and someone's response to us, an otherwise Type 2 NAT device may appear to occasionally behave like a Type 3 NAT device.
Another common gotcha is that some routers which behave as Type 2 when there's just one client using the private network begin to behave as Type 3 when there are two clients both using the same private port numbers on that private network.
There is no "standard" implementation of NAT; everybody started from the basic concept for what a NAT device should do, but implemented it separately, so the little details are typically different in every device, and our "NAT types" discussion is just a useful classification system for carving useful information out of a broad spectrum, for the purposes of establishing peer to peer communication when both peers are behind different NAT devices.
Just something to bear in mind.
On the technical support query in the comment
On the PlayStation, NAT type 2 is "moderate", not "restricted" (not sure where you got that term from). It should work perfectly well, since all modern game services are built to perform standard hole-punching to allow clients to communicate with each other.
How to classify a NAT device's behaviour, yourself
To classify a router between the three types we're discussing here, all you need is a server with three different public IP addresses. We'll call these IP addresses 'A', 'B', and 'C'. Follow these steps:
Client sends a packet to A, and a packet to B.
Server compares the packets it received on its A and B addresses, and checks to see if the public IP address and port numbers it sees as the source of those two packets match each other. If not, the client is Type 3 (Strict).
If the packets did both show the same public source IP address and port, then the server sends a packet from C to that IP address and port. Remember that the client has never sent a packet to C, so if the client successfully receives that packet, then we know it's Type 1 (Open). If it fails to receive the packet, it's Type 2 (Moderate).
That's all it takes!
Why any of this matters
Fundamentally, any number of Type 1 and Type 2 clients can successfully talk to each other; the Type 1 clients because anybody can talk to them on their public address/port, and the Type 2 clients because they can hole-punch out, as long as they know the public address/port that is going to be contacting them.
Type 3 clients, on the other hand, can only talk to Type 1 clients, and they require the Type 1 client to have special NAT traversal code to do that successfully; (the program has to be smart enough for the Type 1 client to receive a packet it wasn't expecting from an unknown address/port, and realise that it must actually be the true address/port from the Type 3 client, and redirect itself to talk to that address, instead of the one it had been expecting to hear from. I would be surprised if many folks actually write that sort of code any more; the vast majority of people are using Type 2 NAT devices, these days, so it'd be hard to test and few folks would get any benefit from this work).
Anyhow, Type 3 clients can't talk to Type 2 clients, because the Type 2 clients don't know the correct address/port to send packets to in order to hole-punch through their own NAT device. And obviously two Type 3 clients will likely never manage to reach each other at all, since neither actually knows the correct address/port for talking to the other.
If you're making a multiplayer game, it can be useful to know which NAT type each player is; for example, knowing that Type 3 players will only be able to talk to Type 1 players, you immediately know not to even bother suggesting that they join a game session which includes any Type 2 or Type 3 players; you already know they won't be able to connect to each other!
And similarly, you know not to send Type 2 players to join a game session which includes any Type 3 players; you know they won't be able to connect. All this logic lets you reduce frustrating "trying to connect to game session" processes which will never complete.
And that's pretty much all there is to it!
A:
Your question is a bit fuzzy regarding the terminology, but if i get you correctly:
A multiplayer solution may want to establish both a persistent connection (TCP or a kept-alive UDP) to a well known public server (a matchmaking/handshaking/hole throughpunching server - maybe it has a better name), while at the same time it may want to allow peers to establish direct connections between each other. Game manufacturers want this to work with the common, normal behaviour of popular home routers. By default, a router does not have static NAT behaviour, as it has no pre-set information about the local computers connecting to it. Meaning that the router itself decides what public port number it assigns to a specific local computer, of multiple, in the local network. This port then points at a specific IP address in the local 192.168.n.n address space.
Clients normally announce themselves (tell their home routers public IP + port that points at a local box) to the matchmaking server, which distributes that endpoint information to other clients, after which all clients can begin to bombard each other with packets - this opens the NATs as needed in the entire cloud. Then there may be different behaviours among the NATs (routers), ie. different restrictions regarding how to allow/disallow ports in opposite peers. The worst case is the symmetric NAT, which changes your routers public port per connection, causing the information it announced to a matchmaking server to be useless for another peer, as these are different connections.
The router needs to have freedom to do this, and nobody wants to set up port assignment tables in the routers, most even don't know how to do this. Hence the router will want to use it's entire public port space, as needed, and dynamically. Eventually, the application(s) started inside the local network determine the extent of all this.
Looking from a coding point of view, this can be a matter of convenience. On can use a single socket for talking with multiple peers, in the case of UDP packets. However one must then parse the remote endpoint in each incoming packet and act accordingly. Another alternative is to use an individual socket for each remote endpoint, in which case the OS sends a callback directly to the correct, assigned callback function. An individual socket however requires an individual local port number, which will translate into an individual port number from the router outwards too - again requiring the routers automation to handle this gracefully.
|
{
"pile_set_name": "StackExchange"
}
|
TRPM7 is involved in angiotensin II induced cardiac fibrosis development by mediating calcium and magnesium influx.
Cardiac fibrosis is involved in a lot of cardiovascular pathological processes. Cardiac fibrosis can block conduction, cause hypoxia, strengthen myocardial stiffness, create electrical heterogeneity, and hamper systolic ejection, which is associated with the development of arrhythmia, heart failure and sudden cardiac death. Besides the initial stimulating factors, the cardiac fibroblasts (CFs) are the principal responsible cells in the fibrogenesis cascade of events. TRPM7, a member of the TRPM (Melastatin) subfamily, is a non-selective cation channel, which permeates both Ca(2+) and Mg(2+). Here we demonstrated TRPM7 expression in CFs, and 2-APB (TRPM7 inhibitor), inhibited Ang II-induced CTGF, α-SMA expression and CFs proliferation. Besides, knocking down TRPM7 by shRNA, we proved that TRPM7 mediated both calcium and magnesium changes in cardiac fibroblasts which contribute to fibrosis progress. This study suggested that TRPM7 should play a pivotal role in cardiac fibroblast functions associated to cardiac fibrosis development.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
This site uses cookies and other tracking technologies to provide you with our services, enhance the performance and functionality of our services, analyze the use of our products and services, and assist with our advertising and marketing efforts.
Gas-shielded Metal Arc Welding (MIG/MAG or GMAW)
With MIG (Metal Inert Gas) or MAG (Metal Active Gas) welding - also called Gas-shielded Metal Arc Welding (GMAW) - an arc is maintained between a continuous solid wire electrode and the workpiece. The arc and weld pool are shielded by a stream of inert or active gas. The process is suitable for most materials and filler wires are available for a wide range of metals.
MIG/MAG - GMAW is a rewarding welding method to automate. At ESAB we have many years' experience of this and over the years we have been the forerunner within robotic welding as well as other applications. Our program for automated MIG/MAG - GMAW consists of everything from the simple welding tractor that holds the welding torch for the welder during long and tiring shifts, up to fully-automatic welding applications.
|
{
"pile_set_name": "Pile-CC"
}
|
BALTIMORE (WJZ)– The details in the murder of 8-month-old Reese Bowman are only getting worse.
The baby was assaulted and killed during nap time at her Baltimore daycare Tuesday, just blocks from Little Italy.
RELATED: Police: Infant Dies After Being ‘Tortured’ By Daycare Worker
Police say Leah Walden was constantly looking around as she tortured the baby. She smothered her and slammed her in her crib.
“We’re just heartbroken right now and I don’t have any comments right now,” says a member of the family who owns Rocket Tiers Learning Center.
[Reporter: Did you guys have any idea that something like this was taking place?]
“Absolutely not, never in a million years would we have thought,” says the family member.
The story has rocked all of Baltimore, 23-year-old Walden, a caretaker at the center is accused of “torturing the infant.”
“What can an 8-month-old baby do to you, to make you want to do that?” says Calvin Anderson, who works at True Masters Barbershop across the street from daycare.
Walden told police she found Bowman unresponsive in her crib, but video footage from inside showed a different story.
New charging documents state it captured Walden in a room with the baby who was “clearly awake” and that she “pressed down on a pillow and blanket” over the baby’s face.”While the child kicked its legs” then “violently snatched the baby, slammed her down in the crib” and “violently shook her body.”
“Reese Bowman in my opinion was tortured,” says Col. Stanley Branford of Baltimore City Police.
Since the murder, Rocket Tiers has been shut down by the Maryland State Department of Education and placed under an emergency suspension.
Police say Walden, who worked at the daycare for more than two years, was the only one who had access to the child.
“We have no reason to believe at this time that there was anything else going on besides this particular incident,” Col. Branford says.
Police say the daycare has been extremely cooperative.
The State Department of Education says the center cannot reopen unless the owners file an appeal and an administrative law judge approves it.
Walden is being held without bond.
Police don’t believe there are other victims but are urging any parents with concerns to call the Child Abuse Center at 443-984-7378.
Follow @CBSBaltimore on Twitter and like WJZ-TV | CBS Baltimore on Facebook
|
{
"pile_set_name": "OpenWebText2"
}
|
Background
==========
Lesion resection is often used to remove mandible lesions in the clinical setting, but results in a large-area mandible defect. A surgical approach is used to reconstruct mandibular bony defects. There are at least three different surgical approaches that can be used for reconstructing mandible defects, including free fibula flap, local flap, and bone plates with screws. During reconstruction by free fibula flap, skin and a portion of fibula in the adjoining soft tissue are harvested when the fibula is removed, and the fibula and soft tissues are used to restore the defect by microscopic anastomosis surgery to allow integration with the arteries and veins of the neck \[[@B1]\]. At present, this method is the most prevalent for mandible reconstruction surgery, because the length of the fibula is adequate for the reconstruction of various mandible defects, and in addition the survival rate is up to 90% \[[@B2]-[@B4]\]. Traditionally, surgeons harvest an amount of bone and a proper amount of adjoining soft tissues sufficient to restore the defect area. The fibula is then cut into several pieces, and the small pieces of fibula are arranged to restore the mandibular bony defect and fixed by plates and screws. However, the shape of mandible is resemblance to a circular arc, which is quite different from the geometrically linear shape of the fibula. How to cut the fibular bone into pieces to pile up the original mandible shape is a big problem for doctors. Therefore, it is often the case that reconstruction results in an asymmetric shape of the mandibular bone. This not only causes a cosmetic problem, but also leads to a greater challenge in terms of the design of dental implants and dentures in the future.
Several methods have been proposed in the literature to solve the above-mentioned problems. Due to recent progression in technology, computer assisted surgical planning has become prevalent. Heather et al. \[[@B5]\] applied Virtual surgical planning and Surgical design and simulation to real five cases study with results of obvious benefits. The application of imaging analysis of the data on the three-dimensional orientation of the bone segments during the preoperative planning, which can increase the treatment efficiency \[[@B6]\]. Hallerman et al. \[[@B7]\] and Yeung et al. \[[@B8]\] demonstrated that reverse engineering can be used to construct a mandible digital model and rapid prototyping can promptly generate a mandible model as the preoperative reference. The preplating technique proposed by Marchetti et al. \[[@B9]\] and Ciocca et al. \[[@B10]\] are useful methods for oromandibular reconstruction that can result in good occlusal and functional rehabilitation. Other studies have also reported the preparation of a customized titanium plate for the surgical area before tumor resection, and once the tumor has been removed, the plate can be used as a template for arranging the segments of fibula \[[@B11]-[@B13]\]. Rohner et al. \[[@B14]\] and Kernan et al. \[[@B15]\] suggested designing a cutting template based on the occlusal position, which allows the doctor to cut the fibula along the template. However, as the cutting template needs to be prepared before the operation, it is difficult to modify if any changes are required during the operation.
Among the above-described methods, application of the medical imaging method can reconstruct a 3D digital model and provide a reference model of the mandible before surgery. However, it does not generate a template as a reference for the fibula, and clinicians need to measure manually the amount of bone required and then consider how to cut and arrange the fibula segments. This often prolongs the process of preoperative preparation and might cause errors due to the lower accuracy of manual measurement. A premade cutting fixture is often not able to fit the defect area due to lesion resection being difficult to predict. Many fixtures cannot be manufactured and applied in a timely fashion during surgery. Application of a metal bone plate provides a good reconstruction target for the mandible defect, but it cannot function effectively as a template for fibula cutting and as a positioning tool for reconstruction.
To overcome the above-mentioned deficiencies, this study first established digital models of the mandible and fibula with the corresponding nodes (locations) for cutting planes (CPs) according to the medical images. The simulation design for the lesion resection was created on the basis of the sequential CP point group. The CAMRP \[[@B16],[@B17]\] generated by our research group was used to calculate the optimal lengths and number of fibula pieces, based on a digital mandible model for defect resection and reconstruction using the fibula bone. Design of the cutting jig and the registering fixture were rapidly generated according to the preoperative digital model, and the designed cutting jig has many guide slots which could be chosen if the resection conditions changed during the procedure. The digital models for all the design models, including cutting jigs and registered fixtures, can then be simultaneously made using additive manufacturing (AM) technology prior to the operation. This provided all the necessary requirements for every step of the mandible defect reconstruction process. This study established a technique for lesion resection that can simultaneously complete fibula cutting and mandible defect reconstruction surgery in one operation. The technique can significantly shorten the duration of surgery, reduce the number of the visits for patients, and enhance the accuracy and success rate of the surgery.
Methods
=======
The flow chart of the study was as shown in Figure [1](#F1){ref-type="fig"}, which includes five main processes: model digitizing and parameterization, pre-surgical planning and design, cutting fixture design, cutting fixture rapid manufacturing, and assembly and verification. Below is an explanation of each process.
{#F1}
Digitized and parameterized models of the mandible and fibula
-------------------------------------------------------------
In this study, the digital model of lesion mandible was designed and captured from a skull model which was obtained by 3D tomography from a healthy man. The 3D tomography images which came from an adults sponsor in our previous study approved by IRB (EMRP01099N) of E-DA hospital Taiwan. The data of skull were then manipulated using MIMICS software (version 13.0) to generate a stereolithography (STL) digital file (Figure [2](#F2){ref-type="fig"}A, left). In order to generate a mandible lesion area, 3Matics software was used to design the lesion area by removing the teeth on the model, which created a mandible digital model (Figure [2](#F2){ref-type="fig"}A, right). In addition, a digital model of the fibula (Figure [2](#F2){ref-type="fig"}B) was captured from an artificial teaching sample using a high-resolution scanner (Germany, ATOS 3D SCANNER).
{#F2}
As the data retrieved from imaging scanning were of a random point cloud, as shown in Figure [2](#F2){ref-type="fig"}C, we used a process to parameterize non-ordered data in the scanned image. The process used Surfacer (version10.0), a reverse engineering software, to calculate the middle lines of the projected outline from the top view and front view of the mandible (Figure [3](#F3){ref-type="fig"}A and B). The middle line planes were obtained from the middle lines of the top view and front view of the mandible, and an intersecting line of the two middle line planes was generated (Figure [3](#F3){ref-type="fig"}C). The intersecting line is defined as the reference central line of the mandible, which is called the "center line" for short.
{#F3}
The center line was partitioned into 100 isometric segments. The point coordinates of all the segments were calculated and 101 node coordinates were obtained in total (Figure [3](#F3){ref-type="fig"}D). According to the 101 nodes of the mandible center line, the normal vector plane going through all the nodes, which was defined as the CP, was calculated. Then, the intersection of the CP and the mandible STL model outline was calculated to gain the points cloud of 101 CPs of the mandible, and according to the points cloud group of these 101 CPs, a parameterized mandible digital model (Figure [4](#F4){ref-type="fig"}A) was reconstructed. Thus, this parameterized model was combined from the point cloud of the 101 CPs. According to the CPs going through every node, cutting calculation of the mandible can be carried out. The same method was used to construct the parameterized digital model of the fibula. Figure [4](#F4){ref-type="fig"}B shows the parameterized digital model of the fibula with clear node CPs. Creating a model with sequenced nodes could improve the efficiency of the cutting simulation.
{#F4}
Pre-surgical planning and shape design
--------------------------------------
By using the parameterized mandible model, the lesion area to be removed was planned based on the planes corresponding to each numbered node. An optimal program was used to calculate the cutting length, the number of the fibula pieces required and the best arrangement under different lesion sizes. Figure [5](#F5){ref-type="fig"}A shows the results calculated by the optimal program, in which 3 pieces of fibula were arranged in the defect area under different defect conditions. Figure [5](#F5){ref-type="fig"}B shows the designs of the mandible reconstruction digital models that contain 3 pieces of fibula corresponding to the different defect conditions shown in Figure [5](#F5){ref-type="fig"}A.
{#F5}
Design of the cutting fixture and registered fixture
----------------------------------------------------
Using the design of the optimal program shown in Figure [5](#F5){ref-type="fig"}, the cutting fixture for the mandible lesion and the registered fixture for the arrangement of fibula pieces were created to improve the surgical accuracy and efficiency. As a perfect match between the mandible cutting fixture and the surface of the mandible is critical, this study retrieved the points cloud data of the mandible defect area and utilized the data to reconstruct the pad of the cutting fixture. Therefore, the fixture can completely fit onto the surface of the mandible. In order to deal with possible variation of the lesion area found during surgery, the mandible cutting fixture was designed with three cutting grooves, which allowed the cutting to be more flexible when a larger or smaller lesion area was found during the operation.
The design of the cutting fixture was made according to the required cutting length and the number of pieces of fibula, and the cutting face was vertical to the center line of the fibula. The design can make the cutting simple to perform, so fibula cutting can be completed in a short amount of time. Figure [6](#F6){ref-type="fig"} shows the design of the cutting fixtures of the fibula and mandible, in which Figure [6](#F6){ref-type="fig"}A is the fibula cutting fixture that presets the cutting length according to the different cutting lengths and positions of the pieces of fibula. Figure [6](#F6){ref-type="fig"}B show the 3D designs of the mandible cutting fixtures. As the internal surface of the cutting fixture perfectly matches the surface of the mandible, the cutting fixture can quickly be fitted onto the position of the mandible lesion to be removed. Figure [6](#F6){ref-type="fig"}B show the designs of fixtures with 2--3 cutter grooves, which could be adapted to the needs of flexible resection if the area of the mandible defect is larger or smaller than postoperatively predicted.
{#F6}
After completion of the cutting of the mandible lesion model and fibula, the pieces of the fibula need to be arranged and positioned into the space between the two cutting ends of the mandible defect. In this study, a registered fixture was designed using Boolean algebra to calculate the optimal combination of fibula pieces. The registered fixture can help to rapidly arrange the fibula pieces onto the defect area (Figure [5](#F5){ref-type="fig"}), satisfying the clinical needs for mandible defect reconstruction. Figure [6](#F6){ref-type="fig"}C shows the registered fixtures for fibula arrangements adapted for 3 pieces of lesion conditions.
Application of Additive Manufacturing (AM) for cutting of the entity
--------------------------------------------------------------------
The completed mandible model, fibula model, cutting fixture and registered fixture are digital models with complicated geometric characteristics. AM technology was applied to rapidly output the entities of the cutting fixture and registered fixture to meet the customized requirements for clinical application in a timely manner. The mandible cutting fixture can be used to fit onto the mandible for rapid and accurate cutting in the clinic, and the registered fixture can ensure that the cut fibula pieces are rapidly positioned and arranged, which can allow mandible reconstruction surgery to be completed in a short time.
Verification and assembly
-------------------------
The error values of the design files for the cutting fixture and registered fixture were verified using reverse engineering software. To ensure the accuracy of the model, the error of the digital model was controlled to within 0.1 mm. and the physical model of fixture was controlled to within 0.5 mm. The fibula entity, after being output by the AM system, was cut using the fibula cutting fixture. The cut fibula pieces were installed onto the registered fixture and then placed onto the defect area of the mandible. The results showed that this technique can achieve the target of mandible reconstruction.
All-in-one mandible reconstruction process
------------------------------------------
After completing the design, cutting planning and registration process, the cutting jig and registered fixture were prepared by AM systems before surgery. Figure [7](#F7){ref-type="fig"} shows the detailed process of cutting, registering, and mandibular reconstruction. First, the defected mandible STL model was input into the CAMRP, and the doctor decided on the location for lesion resection. Then, the CAMRP calculated the optimal length and pieces of fibula for cutting. The custom-made mandible cutting guide-jigs were generated from the neighboring surfaces of the cutting locations. Meanwhile, the number and length of fibular segments were known, and the registering fixture was ready for surgery (Figure [7](#F7){ref-type="fig"}A). The second step was fibula cutting according to the physical guiding fixtures. The harvested fibula bone was set and cut on the alumina alloy fixture based on the length and number of sections according to the planning of the CAMRP (Figure [7](#F7){ref-type="fig"}B). Third, mandibular cutting guide-jigs were snap-fitted, neighboring the lesion area, for resection according to the surgical planning (Figure [7](#F7){ref-type="fig"}C). Finally, the doctor placed the fibula pieces on the registering fixture at the designed position to shape the mandible (Figure [7](#F7){ref-type="fig"}D). If the initial resection was not long enough, the guide-jigs provided some more slots for cutting, and the new cutting location was input into the CAMRP for re-planning of the cutting of the fibular bone. The process of this study, which removes and reconstructs the mandibular lesion using fibular bone segments in one step, is easy to reproduce in clinical practice.
{#F7}
Results
=======
Reverse engineering technology and computer tomography were applied to reconstruct the digital model in this study (as described in Figure [2](#F2){ref-type="fig"}). Figure [6](#F6){ref-type="fig"} shows the design of the cutting fixture and the registered fixture. Figure [8](#F8){ref-type="fig"} shows the detail of the mandible and fibula model entities outputs by the AM system as well the cutting fixture and registered fixture. The mandible and fibula entity models are useful objects for planning, design, discussion and measurement (Figure [8](#F8){ref-type="fig"}A,B). The fibula cutting fixture is shown in Figure [7](#F7){ref-type="fig"}, and Figure [8](#F8){ref-type="fig"}C shows the AM parts of mandible lesion cutting fixture with 2--3 cutter grooves set up on the mandible model, which is used during lesion removal.
{#F8}
The fibula was placed on the fibula cutting fixture to cut it into suitable pieces, and the cut fibula pieces were then arranged on the registered fixture in sequence (see Figure [9](#F9){ref-type="fig"}A). After the position orientation was completed based on registered fixture, the fibula assembly was fixed and available to be quickly set up onto the mandible defect area for completion of the mandible reconstruction entity model. Figure [9](#F9){ref-type="fig"}B shows the AM parts of the mandible reconstruction physical objects that contain 2-4 pieces of AM parts of fibula corresponding to the different defect conditions.
{#F9}
The error analysis of designed digital model was shown as Figure [10](#F10){ref-type="fig"}. Figure [10](#F10){ref-type="fig"}A show the right curved surfaces of the cutting fixture for resection of the mandible lesion area, which is the foundation of the mandible cutting fixture. The design of the benchmark curved surface of the cutting fixture was compared with the tomography point data of the original mandible to clarify the range error. Figure [10](#F10){ref-type="fig"}B are graphs of the range differences between the designed benchmark curved surface of the cutting fixture and the tomography point data of the original mandible. The maximum and average errors of digital model are as shown in Table [1](#T1){ref-type="table"}, which indicates that the maximum error of digital design model was 0.30 mm and the average error of digital model was 0.03 mm. The error map analysis of real model was shown as Figure [11](#F11){ref-type="fig"}. Figure [11](#F11){ref-type="fig"}A show the graphs of the range differences between the digital model of mandible and the scanning point data of the real model of mandible. Figure [11](#F11){ref-type="fig"}B are error map of the range differences between the digital model of cutting fixture and the scanning point data of the real model of cutting fixture. The maximum error and average of real model of mandible and fixture are as shown in Table [2](#T2){ref-type="table"}. The maximum error of real model of mandible was 0.97 mm and average error of real model of mandible was 0.13 mm. The maximum error of real model of cutting fixture was 0.70 mm and average error of real model of mandible was 0.15 mm.
{#F10}
######
Comparison of the errors of the digital model of the mandible cutting fixture
**Item** **Error (Max.)** **Error (Avg.)** **Standard deviation**
------------------------------------------- ------------------ ------------------ ------------------------
Right side cutting fixture curved surface 0.30 mm 0.03 mm 0.06 mm
{#F11}
######
Comparison of the errors of the real model of mandible and cutting fixture
**Item** **Error (Max.)** **Error (Avg.)** **Standard deviation**
----------------------------------- ------------------ ------------------ ------------------------
The real model of mandible 0.97 mm 0.130 mm 0.11 mm
The real model of cutting fixture 0.70 mm 0.15 mm 0.16 mm
Discussion
==========
In traditional mandible reconstruction surgery, several segments of the fibula are harvested from the patient and are normally manually arranged onto the defect area after the mandible lesion is removed. However, due to the shape of the fibula being linear and quite different from the curved shape of the mandible, the reconstructed mandible appearance of the patient is often asymmetric or causes cosmetic issues. In addition, the large difference in the geometric shapes can cause difficulties when dental implants and dentures are installed. In this study, we used advance medical imaging and reverse engineering technology to obtain digital models for patients as the basis of a parameterized digital model, and applied the technique to speed up the design prior to the surgical procedure.
The six steps to fulfill the mandibular reconstruction postulated by this study, which can easily be reproduced and is suitable for clinical practice or other application, are as follows:
1\. Input the patient's 3D mandible model into the CAMRP. The doctor then decides the cutting nodes to remove the lesion portion.
2\. The CAMRP performs surgical planning and generates 3D models of the cutting jigs and fixture.
3\. Output the physical cutting jigs and fixture by 3D printing technology.
4\. Use the guiding jigs to cut the ridus portion of the mandible.
5\. Harvest the fibula, place it on the fixture, and cut into pieces based on the lengths computed by the CAMRP.
6\. Assemble the fibular segments into a registering fixture for mandibular reconstruction.
The periosteum and pedicles are very important in the surgery. In the real cases, we could design a tray under the clamping parts of fibula cutting fixture to support the periosteum and pedicles parts of fibula cutting. The clamping parts will limit the cutting direction, so we need to rotate the fibula bone by centerline of axis to the proper direction where can easy to fulfill the cutting step. This step is very important to avoid of damage of periosteum or pedicles by gravity loading and external force in the surgery process. The design of the fibula cutting fixture should be simplified as much as possible, so as to allow the doctor to rapidly position the fibula and cut it accurately, while the design of the mandible lesion cutting fixture should be as flexible as possible. Therefore, the fixture can be used flexibly, as the lesion might be larger or smaller than predicted, which can only be found when surgery is performed. The digital files of the fibula and mandible cutting fixtures were used to rapidly output the entity fixtures through the AM system. This provided a useful reference for the design of the cutting fixture and registered fixture, which are key in this study. The results of this study demonstrated that our proposed method can help to complete mandible lesion resection and fibula defect reconstruction in one single surgery, and is expected to achieve a good effect in clinical application.
The CP of the fibula cutting fixture was designed to be vertical to the center axis of the fixture. In order to improve the outcome of the arrangement of fibula pieces, the interference at the corner of the bone needs to be removed (overlapped area of Area I in Figure [12](#F12){ref-type="fig"}). Thus, our registered fixture has the function to remove the corner of overlapped bone, which makes clinical mandible defect reconstruction surgery more convenient (groove design for cutting in Figure [7](#F7){ref-type="fig"}). In addition, the removed corner can be used to fill the gap of Area II (Figure [12](#F12){ref-type="fig"}A), which can make the fibula pieces connect better and lead to the reconstruction being more successful. Figure [12](#F12){ref-type="fig"}B and C show the process and completion of the reconstruction model by filling up with the small bone pieces removed from Area I.
{#F12}
A universal design of based surface can be designed for the mandible cutting fixture, which would fit very closely with the mandible surface (Figure [10](#F10){ref-type="fig"}A,B). This surface can be used to design cutting grooves to adapt to the cutting needs of different cases, which can save a lot of time in terms of redesign of the cutting fixtures (Figure [7](#F7){ref-type="fig"}). The design of the cutting fixtures is based on this universal based surface, and the curved surface is designed to stretch over the two sides of the mandible bone ridge, with the curved surface extension depth of the lingual side being controlled to within 2 mm. The curved surface extension depth of the buccal side should not be over half of the front view height of the mandible, thus avoiding failure of installation of the cutting fixture owing to interference.
The results of the Boolean calculation for the fibula optimal assembly can be applied to design and arrange the registered fixture, and also to install the fibula pieces in the defect area. This can help to rapidly position the fibula pieces in an optimal sequence, and so can increase the surgical accuracy and shorten the surgical duration.
Recently, the medical image modeling and AM medical tooling are well service for the planning, design, simulation and manufacturing of pre-surgery and clinical application \[[@B18]-[@B20]\]. Not only the commonplace application of design and simulation of pre-surgery of fibular and mandible cutting surgery, but also we provided the real cutting and setting fixture by AM system which are essential of fulfilling one-stop solution processing of mandible reconstruction of this research. In this study, the fibula cutting fixture was designed from lightweight aluminum, which reduced the weight and increased the convenience for clinical application. Aluminum products could be used repeatedly after autoclave sterilization. The mandible fixtures were made of very strong and elastic nylon material. In terms of practical application, there are more and more material options as the AM system is maturing \[[@B21],[@B22]\]. In recent years, AM technology has been used to output materials that the human body is able to absorb (such as PLA) or materials of human body affinity (such as titanium alloy). Therefore, registered fixtures made of these materials can stay in the human body, and many surgical procedures can be made shorter and simpler.
This study is belonging to simulation of pre-surgery, while not yet enter the process of human subject research. After the processing of digital design, simulation, manufacturing, assembly and testing, the results of this study reveal the high feasibility and effectiveness in the real clinical application. We will plan an advance clinical surgery researching in the near future for the purpose of this study.
Conclusions
===========
In this study, we created the processes of one-stop solution of mandible construction which the surgical processes for patients with oral cancer were simulated, including completion of mandible lesion resection, preparing fibula pieces, recombining the fibula and completion of the mandible defect reconstruction. The designed models show that this advanced method can achieve a high accuracy with the help of a guiding-jig. This reduces the stress on the doctor during surgery, and the surgical procedure can be completed within a shorter time. The application of computer-aided engineering as well as advanced AM technology allows all these processes to be fulfilled in one operation. The technique established in this study can be applied in clinics to greatly reduce the duration of surgery, shorten the course of treatment, and decrease the number of return visits for patients. It can significantly increase the accuracy and success rate of mandible reconstruction surgery, as well as give patients a better facial appearance, as a symmetrical mandible can be easily achieved.
Abbreviations
=============
AM: Additive manufacturing; STL: Stereolithography; CP: Cutting plane.
Competing interests
===================
The authors declare that they have no competing interests.
Authors' contributions
======================
MJT initialized the original concept of this study and the optimal program for the mandible reconstruction using a fibula flap. CTW participated in the design of the study and performed the data analysis. Both authors read and approved the final manuscript.
Acknowledgements
================
The authors want to thank the project support (NSC99-2321-B-006-016) from National Science Council. And the partial financial support from the Medical Device Innovation Center, NCKU is appreciative. The authors would like to acknowledge Dr. Chih-Yu Chen from Dental Department, E-DA Hospital for his valuable suggestions and supports to surgery simulated in the process of assembly and testing.
|
{
"pile_set_name": "PubMed Central"
}
|
Quantum Time Crystals from Hamiltonians with Long-Range Interactions.
Time crystals correspond to a phase of matter where time-translational symmetry (TTS) is broken. Up to date, they are well studied in open quantum systems, where an external drive allows us to break discrete TTS, ultimately leading to Floquet time crystals. At the same time, genuine time crystals for closed quantum systems are believed to be impossible. In this study we propose a form of a Hamiltonian for which the unitary dynamics exhibits the time crystalline behavior and breaks continuous TTS. This is based on a spin-1/2 many-body Hamiltonian which has long-range multispin interactions in the form of spin strings, thus bypassing previously known no-go theorems. We show that quantum time crystals are stable to local perturbations at zero temperature. Finally, we reveal the intrinsic connection between continuous and discrete TTS, thus linking the two realms.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
William Blacklock
William Blacklock may refer to:
William James Blacklock (1816–1858), English landscape painter
William Kay Blacklock (1872–1924), British watercolours and oils artist
William Blacklock House, a historic building in Charleston, South Carolina
|
{
"pile_set_name": "Wikipedia (en)"
}
|
At this point in my life, I go through most days without making a new friend. Sure, an acquaintance or associate here and there, but fewer opportunities to connect with individuals in meaningful, emotional ways. So when I was presented with the chance to build a special friendship with Philadelphia Eagles linebacker Jordan Hicks, I jumped in my car and set forth on a journey neither of us will soon forget.
Jordan exemplifies strength, and not just with those muscles of his. He's an inspiration because of how he's been there for everyone in his life, no matter the circumstance. Combine that with his generosity and genuineness and you've got yourself the foundation for a good friend. I guess you could say it was just my lucky day.
How could I ever prove myself worthy, though? How do I show him what we have in common and what he would get from our friendship? How do I tell him who "Will" is? It may be tough, but if there's a friend at the end of the tunnel, I'll do whatever it takes.
So what do you say, Jordan Hicks: Will you be my friend?
|
{
"pile_set_name": "OpenWebText2"
}
|
Accent designs with earthy shine using the Czech glass 4mm amber with Picasso finish faceted round beads. These beads feature a rounded shape with facets cut into the surface for extra shine. These small beads would make excellent spacers in designs. Each bead features a pale transparent amber color with a mottled Picasso finish in brown. These beads are handmade in small batches, so colors may vary slightly from batch to batch. They are available in 50 piece packs.
|
{
"pile_set_name": "Pile-CC"
}
|
4-Aminobutyrate aminotransferase (GABA-transaminase) deficiency.
4-Aminobutyrate aminotransferase (GABA-transaminase, GABA-T, EC 2.6.1.19) deficiency (McKusick 137150), an inborn error of GABA degradation, has until now been documented in only a single Flemish child. Compared to the other defects of GABA degradation, succinic semialdehyde dehydrogenase (SSADH, EC 1.2.1.24) deficiency with > 150 patients (McKusick 271980) and pyridoxine-dependent seizures with > 100 patients ('putative' glutamic acid decarboxylase (GAD, EC 4.1.1.15) deficiency; McKusick 266100), GABA-T deficiency is very rare. We present a summary of the clinical, biochemical, enzymatic and molecular findings on the index proband, and a recently identified second patient, with GABA-T deficiency. The phenotype in both included psychomotor retardation, hypotonia, hyperreflexia, lethargy, refractory seizures and electroencephalographic abnormalities. In an effort to elucidate the molecular basis of GABA-T deficiency, we isolated and characterized a 1.5 kb cDNA encoding human GABA-T, in addition to a 41 kb genomic clone which encompassed the GABA-T coding region. Standard methods of cloning and sequencing revealed an A-to-G transition at nucleotide 754 of the coding region in lymphoblast cDNAs derived from the index proband. This mutation resulted in substitution of an invariant arginine at amino acid 220 by lysine. Expression of the mutant in E. coli, followed by isolation and enzymatic characterization of the recombinant protein, revealed an enzyme whose Vmax was reduced to 25% of wild-type activity. The patient and father were heterozygous for this allele; the second allele in the patient remains unidentified. Genomic Southern analysis revealed that the second proband most likely harbours a deletion in the 3' region of the GABA-T gene.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
###### depositTokens
`https://etherscan.io/tx/0xb36ec348dd24a065ddbb765675501d0bb933d2092028c9b03568c9b293a6eab3`
###### withdrawEthers
`https://etherscan.io/tx/0x663a8fd9356dfd35e82f08475fdf5164bac00ffe7ee9510d33ad6d16985a27b1`
###### withdrawTokens
`https://etherscan.io/tx/0x256574e252aee51822f6bd73088cb5611068b3c8aa56d481dad08ba87d351a62`
###### takeSellOrder
`https://etherscan.io/tx/0x2e068d180e015087fe1a67732ae1c0d70b42bd44530b2e96944d68ae13b86088`
`https://etherscan.io/tx/0x777a454e3bf2b8e4d09ca0ffea5b7ec03f596d2baae935f4bbcb4a13a6104890`
###### takeBuyOrder
`https://etherscan.io/tx/0xfed6bd46d7e4bac5698e874b56b658e06aaf5aa1f989294d2551da1f73ccec65`
###### cancelOrder
`https://etherscan.io/tx/0x04d34a02a627f85797a96f36078846ac09e13888c4b7e008013e2cb1b3900a8d`
|
{
"pile_set_name": "Github"
}
|
Twelve tone equal temperament table in hertz
This is a pre-genereated table which shows all octaves and notes in hertz.
|
{
"pile_set_name": "OpenWebText2"
}
|
“Southern Soul” performing artist Rissi Palmer will take to the stage at the Wake Forest Renaissance Centre for the Arts Saturday, May 11, at 7:30 p.m. Advance tickets are on sale for $15 (plus tax) and can be purchased online with a Visa, MasterCard, American Express, or Discover credit card at… Click here to read the full Wake Forest Gazette ... Read More »
ArtsPower National Touring Theatre will offer two performances of “Your Alien” on Friday, April 26, at the Wake Forest Renaissance Centre, 405 Brooks Street. Geared for audiences ages 12 and under, the shows are scheduled for 11 a.m. and 6:30 p.m. ArtsPower’s newest musical touches down with a s… Click here to read the full Wake Forest Gazette story Read More »
The Wake Forest Area Chamber of Commerce Foundation is currently accepting applications for the inaugural class of its Career Readiness Academy. Rising juniors and seniors for the 2019/2020 school year are eligible to apply. Funded through a grant from the US Chamber of Commerce, the goal is to p… Click here to read the full Wake Forest Gazette story Read More »
The Wake Forest Technology Advisory Board is inviting area middle and high school students and their parents to Wake Forest STEM Saturday 2019 on Saturday, April 13. Presented by the Wake Forest Wireless Research Center, this free family networking fair is scheduled from 10 a.m to 2 p.m. at the W… Click here to read the full Wake Forest Gazette ... Read More »
As someone who has used an absentee ballot for voting in the last two elections, I was heartened to see that the State Board of Elections is considering requiring prepaid envelopes for absentee ballots. The ballot and the envelope are large – the ballot was 17 inches long last year – and the enve… Click here to read the full ... Read More »
On Mondays at 6 p.m. from March 25 through April 29 local people are invited to the Alston-Massenburg Center at 416 North Taylor Street for a program called Beyond Bites, six weeks of free cooking classes to help them make better nutritional decisions, save money on groceries and improve your ove… Click here to read the full Wake Forest Gazette ... Read More »
The Wake Forest Inspections Department issued permits for 63 new homes during February – 51 for single-family homes and 12 for townhouses. Inspectors also provided inspections for fit-ups and pre-occupancy for seven new business: Animal Eye Care at 1608 Heritage Commerce Court, Suite 160; South … Click here to read the full Wake Forest Gazette story Read More »
Wake Forest Issue Reported on Wed, 20 Mar 2019 12:42:59 -0400: Trash/Debris on Public Property at Address: 907 Amersham Lane Wake Forest, North Carolina Rating: 1 907 Amersham Lane has left a large chair and two mattresses on the curb in front of their house. They seemed to have moved out. This house is at the front of the neighborhood ... Read More »
|
{
"pile_set_name": "Pile-CC"
}
|
Guillaume Bouzignac
Guillaume Bouzignac (c. 1587 – c. 1643) was a French composer.
Bouzignac was probably born in 1587 in Saint-Nazaire-d'Aude. He studied at the Cathedral of Narbonne until 1604, and was choirmaster at the Cathedrals of Angoulême, Bourges, Tours, and Clermont-Ferrand.
His motets are preserved in two manuscripts. His motets are highly distinctive: "Simply stated, there is no other music of the time that looks the same on the page or sounds the same as the motets of Bouzignac." "One name in this period rises above those of his contemporaries for all sacred music, including Masses: that of Guillaume Bouzignac." His dialogue motets, such as Unus ex vobis and Dum silentium, are small scale oratorios which anticipate Giacomo Carissimi, and then Marc-Antoine Charpentier (1643–1704) two generations later.
Discography
Motets, Te Deum Les Arts Florissants dir. William Christie Harmonia Mundi HMC 901 471
Motets with Jean Gilles Leçons de Ténèbres Boston Camerata dir. Joel Cohen Erato 4509 98480-2
Motets Sächsisches Vocalensemble dir. Matthias Jung TACET S 156 2007
References
Category:1587 births
Category:1643 deaths
Category:French Baroque composers
Category:French classical composers
Category:French male classical composers
Category:People from Aude
Category:17th-century classical composers
Category:17th-century French composers
|
{
"pile_set_name": "Wikipedia (en)"
}
|
Reserve Bank of India ( RBI ) deputy governor NS Vishwanathan said the job of banks is not to bail out borrowers whenever they get into difficulties as the primary commitment of lenders should be to depositors.Vishwanathan’s address staked out the position of the central bank on this and other issues, some of them in apparent contrast with the government’s views, and followed comments by fellow deputy governor Viral Acharya on October 26 that had sparked a swift retort from the finance minister.“Banks are not supposed to be shock absorbers of first resort of the difficulties faced by their borrowers as banks do not have the luxury of delaying payments to their depositors,” Vishwanathan said in his October 29 speech in Jamshedpur that was posted on the RBI website late on Friday. His comments come against the backdrop of non-banking finance companies (NBFCs), in the grip of a liquidity squeeze seeking a rescue from banks. This is one of the issues on which the government is said to differ with RBI, which is opposed to a proposed special liquidity window for NBFCs.Vishwanathan also defended RBI’s February 12 circular, which tightened loan-recognition norms and scrapped all existing debt-restructuring schemes for large borrowers since the Insolvency and Bankruptcy Code (IBC) had been put in place.Of course, a bank can renegotiate terms of a loan if circumstances warrant, but this must be for a good reason and the bank should recognise the consequent risks,” Vishwanathan said. “This renegotiation of terms should be an exception rather than the rule, as resorting to it often would endanger the safety of deposits, dent a bank’s ability to lend further and imperil its existence as an intermediating entity.”There had been a demand from the power sector for special dispensation in restructuring existing debt, a position that the government had backed in court.“Next time we hear about a bank making efforts to recover loans from borrowers, we should all note to remember it is essentially trying to get back the depositors’ money,” the deputy governor said.The most important objective of the central bank’s Revised Framework for Resolution of Stressed Assets is to alter the balance of power in favour of creditors, something which had been in favour of debtors for long time, he said.“The out-of-court restructuring mechanisms too suffered high failure rates resulting in the borrowing entities continuing to indulge in repeated defaults, being confident the balance of power remained with them and the ability of banks to discipline errant borrowers was weak,” the deputy governor said, explaining why the IBC process was necessary.In the case of wilful or strategic defaulters, who can pay but won’t do so a change in ownership accompanied by “punitive action against the defaulting management is the way to go”, he said.“The Indian banking system has a high proportion of un-provided NPAs vis-àvis the capital levels. As I said, there are signs of improvement in the default rates and recovery rates after IBC and RBI’s Revised Framework, which may result in lower unexpected losses for banks in the future,” Vishwanathan said. “However, a recalibration of risk-weights or minimum capital requirements would need to wait till these trends are firmly entrenched in the economy. Frontloading of regulatory relaxations before the structural reforms fully set in could be detrimental to the interests of the economy.”He said that resolution plans offered by existing managements were less valuable than they seemed. “The choice before banks is: ‘illusory future payments’ vs ‘upfront real cash’. Banks need to arrive at the present value of ‘illusory future payments’ by discounting it for time value of money and more importantly for the uncertainty in receiving the payments taking into account the existing management’s past records,” he said.He also spoke about RBI insisting on stricter standards than Basel norms, referring to complaints that capital requirements for banks are unnecessarily high at 9% rather than 8% under the international guideline. “Regulatory capital is meant to serve as a buffer against unexpected loss,” he said.“The suggestion by some that our capital requirements are more onerous than international standards is not correct at all. As the need for repeated recapitalisation has proved, banks in India need to aspire to have higher capital levels.”He also said that higher CRAR doesn’t necessarily squeeze credit growth but raises hygiene.“It may be noticed that in the past, high levels of credit growth due to ‘supply push’ have resulted in high corporate leverage and consequent NPAs in the banking system,” the deputy governor said. “We must guard against any push for dilution of standards in the name of aligning them with international benchmarks because that will be cherry-picking and will result in our banks being strong in a make-believe sense and not in reality,” Vishwanathan said. “It is by resisting such temptations, I believe, we will build a financial system that is lot stronger than today.”As the insolvency and bankruptcy regime matures, debt recovery and asset quality will get closer to global standards, he said. “Then our probability of default and loss given default will also come down to global levels. Hopefully, those days are nearer than we think,” the deputy governor said.
|
{
"pile_set_name": "OpenWebText2"
}
|
The main objective will be to ensure physical delivery to plan within a Logistics Team striving to continuously add customer value whilst constantly reducing cost.
DUTIES INVOLVED IN THIS VACANCY: *Creates meaning and challenge for team members in the workplace.*Meets daily with the whole team to review performance and agree targets.*Involves the team in decision making and problem solving.*Leads effective, focussed and productive Team Meetings.*Gives clear direction (who, what, when, why, how) when the team needs it.*Empowers the team to perform independently and take responsibility.*Continuously seeks ways to improve performance, is open to and adopts new ideas.*Provides the team with all resources to perform well.*Displays a "can-do" attitude and enthusiasm.*Demonstrates commitment to the organisation, its mission, vision, values and goals.*Conducts oneself at all times in a professional manner to promote a positive image of the Logistics function and UK Power Networks in general, demonstrating the Vision and Values of the company*Takes ownership of customer enquiries and/or concerns originating from the Customer Issue Log, provides efficient feedback and resolution within agreed timeframe targets*Ensures procedures, process flows and SOPs are reviewed and maintained *Completes ad hoc duties as requested by the Operations Manager*Assists with supervision of the Logistics Warehouse and Deliveries Operation from receipt of goods to final delivery, ensures completion of all tasks on time and in full.
THE CANDIDATE: *Candidate must have strong working experience of SAP WMS, *Strong man management skills*Experience of LEAN / Six Sigma continuous improvement. *Experience of managing in both warehouse and transport operations is desirable as the successful candidate would need to be involved in both and ideally with previous experience in industrial / utilities customers and materials.
The first stage of the application process is to apply online. If you would like any more information about this vacancy before applying, please feel free to contact us and ask for John. If you have not been contacted by telephone or email within 7 days of your application, you have been unsuccessful on this occasion. Unless you notify us otherwise, we will retain your CV and covering letter on our database and may contact you with other job opportunities in the future.
|
{
"pile_set_name": "Pile-CC"
}
|
Kerstin Jönßon
Kerstin Jönßon (born 3 July 1964) is a German handball player who played for the West German national team and later for Austria. She was born in Glückstadt. She represented West Germany at the 1984 Summer Olympics in Los Angeles, where the West German team placed fourth.
She represented Austria at the 1992 Summer Olympics in Barcelona.
References
Category:1964 births
Category:Living people
Category:Sportspeople from Schleswig-Holstein
Category:German female handball players
Category:Olympic handball players of West Germany
Category:Handball players at the 1984 Summer Olympics
Category:Austrian female handball players
Category:Olympic handball players of Austria
Category:Handball players at the 1992 Summer Olympics
|
{
"pile_set_name": "Wikipedia (en)"
}
|
/**
* Licensed to jclouds, Inc. (jclouds) under one or more
* contributor license agreements. See the NOTICE file
* distributed with this work for additional information
* regarding copyright ownership. jclouds licenses this file
* to you under the Apache License, Version 2.0 (the
* "License"); you may not use this file except in compliance
* with the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/
package org.jclouds.http;
/**
* Indicate whether a request should be retried after a server error response (HTTP status code >=
* 500) based on the request's replayable status and the number of attempts already performed.
*
* @author James Murty
*/
public interface HttpRetryHandler {
public static final HttpRetryHandler ALWAYS_RETRY = new HttpRetryHandler() {
public boolean shouldRetryRequest(HttpCommand command, HttpResponse response) {
return true;
}
@Override
public String toString() {
return "ALWAYS_RETRY";
}
};
public static final HttpRetryHandler NEVER_RETRY = new HttpRetryHandler() {
public boolean shouldRetryRequest(HttpCommand command, HttpResponse response) {
return false;
}
@Override
public String toString() {
return "NEVER_RETRY";
}
};
/**
* Return true if the command should be retried. This method should only be invoked when the
* response has failed with a HTTP 5xx error indicating a server-side error.
*/
boolean shouldRetryRequest(HttpCommand command, HttpResponse response);
}
|
{
"pile_set_name": "Github"
}
|
I don't know about you, but I love solar eclipses. Only problem is, you can't look at them directly—or shouldn't, anyway. There are none of those rules with Cauca's Eclipsion clock though, which mimics solar eclipses with the positioning of the "hands." The clock costs $80, with $18 of shipping on top too. [Cauca via DesignMilk]
|
{
"pile_set_name": "Pile-CC"
}
|
---
abstract: |
The study of transport and mixing processes in dynamical systems is particularly important for the analysis of mathematical models of physical systems. We propose a novel, direct geometric method to identify subsets of phase space that remain strongly coherent over a finite time duration. This new method is based on a dynamic extension of classical (static) isoperimetric problems; the latter are concerned with identifying submanifolds with the smallest boundary size relative to their volume.
The present work introduces *dynamic* isoperimetric problems; the study of sets with small boundary size relative to volume *as they are evolved by a general dynamical system*. We formulate and prove dynamic versions of the fundamental (static) isoperimetric (in)equalities; a dynamic Federer-Fleming theorem and a dynamic Cheeger inequality. We introduce a new dynamic Laplace operator and describe a computational method to identify coherent sets based on eigenfunctions of the dynamic Laplacian.
Our results include formal mathematical statements concerning geometric properties of finite-time coherent sets, whose boundaries can be regarded as Lagrangian coherent structures. The computational advantages of our new approach are a well-separated spectrum for the dynamic Laplacian, and flexibility in appropriate numerical approximation methods. Finally, we demonstrate that the dynamic Laplace operator can be realised as a zero-diffusion limit of a newly advanced probabilistic transfer operator method [@F13] for finding coherent sets, which is based on small diffusion. Thus, the present approach sits naturally alongside the probabilistic approach [@F13], and adds a formal geometric interpretation.
author:
- |
Gary Froyland\
\
School of Mathematics and Statistics\
University of New South Wales\
Sydney NSW 2052, Australia
title: |
Dynamic isoperimetry and the\
geometry of Lagrangian coherent structures
---
¶
\[section\] \[theorem\][Lemma]{} \[theorem\][Corollary]{} \[theorem\][Sublemma]{} \[theorem\][Remark]{}
Introduction
============
The study of Lagrangian coherent structures in nonlinear dynamics is broadly concerned with the identification of spatial structures in phase space that behave in a relatively “stable” way under the dynamics by resisting high levels of distortion and/or diffusion. In the case of purely advective dynamics governed by a nonlinear map or time-dependent ordinary differential equations, if the structure is a full-dimensional set[^1], this set resists filamentation under the nonlinear dynamics and the ratio of boundary size to the volume of the set remains relatively unchanged. When the dynamics is a combination of advection and diffusion, for example, a time-dependent Fokker-Planck equation, a coherent set resists mixing with the surrounding phase space, again through the mechanism of retaining a relatively low boundary size. There is a long history of development of related ideas spread across the dynamical systems, fluid dynamics, and geophysics literature. We mention just two early related works: [@pierrehumbert_yang], which contains several ideas concerning mixing and transport mitigation in fluids, and the book [@ottino], which discusses purely advective (chaotic) mixing. These ideas, and the theory and algorithms developed subsequently, have grown into their own field, and have been employed across a wide spectrum of physical, biological, environmental, and engineering applications. The study of sets with minimal boundary is known in differential geometry as an isoperimetric problem. The classic isoperimetric problem in $\mathbb{R}^2$ is to determine the set $S$ with least boundary length (perimeter), given a fixed area; or equivalently to find a set $S$ of fixed (iso-) perimeter with greatest area. The unique solution of this problem is a disk; all other sets $S$ satisfy the inequality $\mbox{length}(\partial S)/\sqrt{\mbox{area}(S)}>2\sqrt{\pi}$, an example of an *isoperimetric inequality*. The obvious generalisation of this problem in $\mathbb{R}^d$ is true: $d$-balls minimise surface area and $\ell_{d-1}(\partial S)/\ell_d(S)^{1-1/d}>d{\omega_d}^{1/d}$, where ${\omega_d}$ is the volume of a unit ball in $\mathbb{R}^d$, and $\ell_d, \ell_{d-1}$ are $d$ and $d-1$-dimensional volume (see e.g. [@chavelisoperimetric]). Because we have in mind applications to fluid flow, we focus on compact domains $M$ rather than $\mathbb{R}^d$. For compact, connected $M$, one has a hypersurface $\Gamma\subset M$ disconnect $M$ into two pieces $M_1,M_2$, just as the $d$-ball disconnects $\mathbb{R}^d$. One tries to find a disconnecting hypersurface $\Gamma$ that minimises the ratio $$\mathbf{h}(\Gamma):=\ell_{d-1}(\Gamma)/\min\{\ell_d(M_1),\ell_d(M_2)\}.$$
One of our main contributions is to develop a theory of *dynamic* isoperimetry, where one studies the *evolution* of hypersurfaces $\Gamma\subset M$ that disconnect phase space $M$, under a nonlinear transformation $T:M\to T(M)$. Both the manifold $M$ and the separating surface $\Gamma$ are subjected to general nonlinear dynamics, representing the action of some (possibly chaotic) flow over some finite-time duration. The solution to this dynamic isoperimetric problem may have nothing to do with the solution to the static problem because even if $\Gamma$ has low co-dimension 1 volume, the size of $T(\Gamma)$ may be greater, and if $T$ is chaotic, $T^k(\Gamma)$ for a modest number of iterations $k$ may have significantly greater size (see Figure \[coherencefig\]b and \[coherencefig\]c).
![The two-dimensional set on the left with boundary $\Gamma$ has a low boundary size to volume ratio. The three sets $T(\Gamma)$ on the right are three possible images of the shape on the left under three different nonlinear volume-preserving dynamical systems $T$ over a fixed finite time duration. Under dynamics ‘a’, the set on the left retains a low boundary size to volume ratio, but under dynamics ‘b’ and ‘c’, the boundary size is significantly increased.[]{data-label="coherencefig"}](coher-noncoher-images_withletters.png "fig:"){width="12cm"}\
Thus, the dynamics plays a key role in the selection of a surface $\Gamma$ that remains small relative to the domain volume when evolved under a general volume-preserving nonlinear dynamical system. Clearly such surfaces $\Gamma$ bound sets that are very natural candidates for [coherent sets]{}. To take a real-world example, coherent sets such as oceanic eddies retain water mass and remain coherent by their boundary remaining as short as possible over an extended period of time. This reduces diffusion through the eddy boundary $T^k\Gamma$ at times $k=0,1,2,\ldots$ via small-scale diffusion processes (e.g. [@balasuriyajones99]).
In discrete time, under a single application of $T$, this motivates the minimisation of the quantity $$\label{cheegereqn_intro}
\mathbf{h}^D(\Gamma):=\frac{\ell_{d-1}(\Gamma)+\ell_{d-1}(T(\Gamma))}{2\min\{\ell(M_1),\ell(M_2)\}},$$ where $\Gamma$ varies over smooth hypersurfaces disconnecting $M$ into two connected pieces $M_1, M_2$. In continuous time, we consider smooth flow maps $T^{(t)}:M\to T^{(t)}(M)$, $t\in[0,\tau]$ and the quantity $$\label{hdynt_intro}
\mathbf{h}_{[0,\tau]}^{D}(\Gamma):=\frac{\int_0^\tau \ell_{d-1}(T^{(t)}\Gamma)\ dt}{\tau\min\{\ell(M_1),\ell(M_2)\}}.$$ In addition to the geometric interpretation of Figure \[coherencefig\], the expression (\[hdynt\_intro\]) is also directly proportional to the mass lost through the boundary over the finite time interval $[0,\tau]$ via continually-present small-scale diffusion. This latter interpretation motivates the additive combination of terms in (\[cheegereqn\_intro\]) and (\[hdynt\_intro\]) (as opposed to e.g. a multiplicative combination). We focus on the difficult setting of general time-dependent dynamics. In the autonomous dynamics setting, classical “coherent” (in fact, invariant) objects such as invariant tori or invariant cylinders are completely invariant under the dynamics and thus their boundaries remain fixed and unchanging. Furthermore, trajectories that begin on the inside of these structures can never leave through their co-dimension 1 boundary. Thus, these objects can be regarded as “ideal” coherent structures and are relatively well-understood. In the general time-dependent dynamics setting, the existence of such perfectly invariant structures is highly unlikely.
\[def:ftcs\] For the finite-time dynamics considered we define a *maximally coherent structure* on $M$ to be a minimizing $\Gamma$ for (\[cheegereqn\_intro\]) or (\[hdynt\_intro\]) (see also Section \[sect:multistep\]) when the infimum is achieved; otherwise we select a $\Gamma$ for which $\mathbf{h}^D(\Gamma)$ is arbitrarily close to the infimum. The corresponding *maximally coherent set* is defined to be the $M_k$, $k=1,2$ with minimal volume arising from the disconnection $\Gamma$.
Existing approaches to identifying coherent structures fall broadly into two categories: probabilistic methods and geometric methods. Probabilistic approaches to finding coherent structures are based around the [transfer operator]{} (or Perron-Frobenius operator) and can be applied to systems with a combination of advection and diffusion, or purely advective dynamics. These methods look for *finite-time coherent sets* [@FSM10; @F13; @FPG14]: sets that resist mixing with the rest of phase space and represent global transport barriers to complete mixing. These constructions have found application in atmospheric dynamics to map and track the Antarctic polar vortex [@FSM10], and in ocean dynamics to track an oceanic eddy in the Agulhas current [@FHRSS12; @FHRvS15], in both two and three dimensions. In the purely advective setting, the constructions underpinning the transfer operator methods rely on small amounts of diffusion [@F13]. This small diffusion makes complete phase space mixing possible and is required for other technical reasons; these points are discussed in [@F13] and in the present work in Section \[sect:zerolimit\]. Further, the boundary sizes of finite-time coherent sets are implicitly measured in [@F13] because the localised diffusion can only eject mass near the boundaries of coherent sets; thus the mixing experienced over a finite time is tied to the boundary sizes of the coherent sets. We therefore expect our proposed dynamic isoperimetry methodology to be compatible with [@F13] and in fact, we show that the former arises as a zero-diffusion limit of the latter, which makes explicit the geometry contained in the probabilistic methods for small diffusion.
In recent years there have been several geometric methods proposed to characterise either trajectories or co-dimension 1 surfaces that represent coherent structures in purely advective dynamics [@mancho_M; @mezicmeso; @Haller_11; @Haller_12; @allshouse_thiffeault; @ma_bollt_shape; @romkedar_M]. In two dimensions, [@Haller_11] defines a hyperbolic LCS as a material curve that has repelling dynamics normal to the curve in forward time and greater repulsion than nearby curves, while [@Haller_12] defines transport barriers as curves that are local minimisers of length functionals integrated over a finite time interval, with various hyperbolic, shear, and elliptic boundary conditions for the associated Euler-Langrange equations. The approach [@ma_bollt_shape] suggests that curves formed from points that experience local rigid-body motion over a finite-time duration are associated with coherent dynamics. Most approaches compute various scalar fields from Lagrangian trajectories and infer corresponding dynamic properties from the fields.
The main contributions of this paper include formulations of dynamic versions of classical objects in isoperimetric theory and formulations and proofs of dynamic versions of fundamental isoperimetric theorems. We formulate a dynamic version of the *Cheeger constant* $\mathbf{h}$ (the minimal ratio of the $d-1$-dimensional volume of a disconnecting hypersurface $\Gamma$ to the disconnected volumes of $M_1, M_2$) and the Sobolev constant (a functional representation of the Cheeger constant, where the separating hypersurface $\Gamma$ is the level set of a smooth function). The celebrated *Federer-Fleming theorem* equates these two constants, formally linking geometric and functional representations of these notions of isoperimetry. We formulate and prove a dynamic version of the Federer-Fleming theorem, linking our new dynamic Cheeger and Sobolev constants (Section \[sect:ff\]).
We further formulate and prove a dynamic version of the *Cheeger inequality*, which relates the second largest eigenvalue of the Laplace operator $\triangle$ on $M$ to the Cheeger constant (Section \[sect:cheegerineq\]). This requires a replacement of the Laplace operator with a new operator that incorporates the general nonlinear dynamics. In the discrete-time volume-preserving setting, the operator on $M$ corresponding to the expression (\[cheegereqn\_intro\]) is $$\label{dynlap_intro}
(\triangle+\mathcal{P}^*\triangle\mathcal{P})/2,$$ where $\mathcal{P}f=f\circ T^{-1}$ is the transfer operator for $T$. We develop a spectral theory for this new operator in Section \[sect:dynlapspec\], and propose an algorithm that uses eigenvectors of this operator to identify coherent sets in practice. In Section \[sect:zerolimit\] we demonstrate that one can recover the new dynamic Laplace operator from the probabilistic methodology of [@F13] as a zero-diffusion limit of the latter. The probabilistic approach in [@F13] computes singular vectors of an $\epsilon$-perturbed operator $\mathcal{L}_\epsilon$; that is, eigenvectors of $\mathcal{L}_\epsilon^*\mathcal{L}_\epsilon$. For $C^3$ $f:M\to\mathbb{R}$ we show that $$\label{diffform1thm_intro}
\lim_{\epsilon\to 0} \frac{(\mathcal{L}_\epsilon^*\mathcal{L}_\epsilon-I)f(x)}{\epsilon^2}=c\cdot(\triangle+\P^*\triangle\P)f(x),$$ for each $x\in \mathring{M}$, where $c$ is an explicit constant. Thus, we provide a missing formal link between the probabilistic coherent set methodologies and direct notions of geometry via boundary size and volume. Finally, in Section \[sect:numerics\], we illustrate how eigenfunctions of the dynamic Laplace operator (\[dynlap\_intro\]) can be used to find coherent sets using three numerical case studies. The appendix contains most of the proofs.
Background {#staticsection}
==========
Let $M$ be a compact, connected $d$-dimensional $C^\infty$ Riemannian manifold of vanishing curvature, which is either boundaryless or has $C^\infty$ boundary. This setting is relatively simple from a differential geometric point of view, but the introduction of nonlinear dynamics creates nontrivial questions in this setting. Let $\ell_d$ denote Lebesgue (volume) measure on $M$. To measure co-dimension 1 objects, we use $d-1$-dimensional Hausdorff measure $\mathcal{H}^{d-1}$ (using the trivial Riemannian metric to calculate diameter) to define $\ell_{d-1}=(\omega_{d-1}/2^{d-1})\mathcal{H}^{d-1}$; see e.g. Corollary IV.1.1 [@chavelisoperimetric]. We define the *Cheeger constant* $$\label{cheegerconst}\mathbf{h}:=\inf_\Gamma\frac{\ell_{d-1}(\Gamma)}{\min\{\ell_d(M_1),\ell_d(M_2)\}},$$ where $\Gamma$ varies over compact $(d-1)$-dimensional $C^\infty$ submanifolds that separate $M$ into two connected components $M_1, M_2$.
One can link these geometric ideas with functions on $M$ by considering level sets of a function $f:M\to\mathbb{R}$ defining the $(d-1)$-dimensional separating surface $\Gamma$. In fact, defining the *Sobolev constant* $$\label{sobolevconst}
\mathbf{s}:=\inf_{f\in C^\infty}\frac{\|\nabla f\|_1}{\inf_{\alpha\in\mathbb{R}}\|f-\alpha\|_1},$$ one has the celebrated Federer-Fleming result
\[ffthm\] $$\label{staticffeqn}
\mathbf{s}=\mathbf{h}.$$
A link to spectral theory of operators is provided by the Cheeger inequality. Consider the eigenproblem $$\label{closedeigenproblem}
\triangle\phi=\lambda\phi,\mbox{ on $\mathring{M}$}.$$ If $\partial M\neq \emptyset$, then zero Neumann boundary conditions are imposed: $$\label{staticneumann}
\nabla \phi(y)\cdot \mathbf{n}(y)=0\mbox{ for $y\in\partial M$},$$ where $\mathbf{n}(y)$ is the outward unit normal to $\partial M$ at $y$.
It is well-known (see e.g. Theorem 1.1 [@chaveleigenvalues] and Section 4.4 [@mcowen]) that the set of eigenvalues consists of a sequence $0= \lambda_1>\lambda_2>\cdots\downarrow -\infty$. and each associated eigenspace is finite-dimensional. Eigenspaces belonging to distinct eigenvalues are orthogonal in $L^2(M)$ and $L^2(M)$ is the direct sum of all of the eigenspaces. Furthermore, each eigenfunction is $C^\infty$ on $M$.
\[cheegerthm\]
1. If $M$ is boundaryless, let $\lambda_2$ be the smallest magnitude nonzero eigenvalue of (\[closedeigenproblem\]).
2. If $\partial M\neq\emptyset$, let $\lambda_2$ be the smallest magnitude nonzero eigenvalue for (\[closedeigenproblem\])–(\[staticneumann\]).
Then $$\label{staticcheegerineq}
\mathbf{h}\le 2\sqrt{-\lambda_2},$$
Level sets of the eigenfunction corresponding to $\lambda_2$ give vital information about the $\Gamma$ that achieves the Cheeger constant; a fact that we will exploit in our new dynamic setup. We remark that there is a vast literature on the use of the Laplace operator for extracting various types of geometric information on static manifolds and we refer the reader to the recent survey [@grebenkov_nguyen], with over 500 references. We now proceed through a few simple domains to illustrate the relationship between the solution to the isoperimetric problems and the eigenvalues and eigenfunctions of the Laplacian.
The flat 2-torus: {#sect:torus}
-----------------
Consider the flat 2-torus $\mathbb{T}^2=2\pi(\mathbb{R}/\mathbb{Z})\times 2\pi(\mathbb{R}/\mathbb{Z})$, which we write as $[0,2\pi)/\sim\times [0,2\pi)/\sim$, where $\sim$ is the identification at the interval endpoints.
### Solution to the isoperimetric problem {#solution-to-the-isoperimetric-problem .unnumbered}
There is an infinite family of optimal $\Gamma$ solving (\[cheegerconst\]): either $(\{x\}\times [0,2\pi))\cup(\{x+\pi\}\times [0,2\pi))$ (two vertical loops) or $([0,2\pi)\times \{y\})\cup([0,2\pi)\times\{y+\pi\})$ (two horizontal loops). The value of $\mathbf{h}$ is $2\cdot(2\pi)/((1/2)\cdot (2\pi)^2)=2/\pi$. One particular solution is shown as black lines in Figure \[torusfig\]: $\Gamma=\{(\{\pi/2\}\times [0,2\pi))\cup(\{3\pi/2\}\times [0,2\pi))$, $M_1=[\pi/2,3\pi/2]\times [0,2\pi)$, and $M_2=\mathbb{T}^2\setminus M_1$.
[0.3]{} ![Cylinder $[0,a)\times [0,b]$, $a=3/2$, $b=1$: Plot of second Laplacian eigenfunction $\sin(\pi y)$ with the the zero level set shown in black.[]{data-label="cylfig"}](torus_cosine_withlines.png "fig:"){width="\textwidth"}\
[0.3]{} ![Cylinder $[0,a)\times [0,b]$, $a=3/2$, $b=1$: Plot of second Laplacian eigenfunction $\sin(\pi y)$ with the the zero level set shown in black.[]{data-label="cylfig"}](rectangle_cosine_withline.png "fig:"){width="\textwidth"}\
[0.3]{} ![Cylinder $[0,a)\times [0,b]$, $a=3/2$, $b=1$: Plot of second Laplacian eigenfunction $\sin(\pi y)$ with the the zero level set shown in black.[]{data-label="cylfig"}](cylinder_cosine_withline.png "fig:"){width="\textwidth"}\
### Laplace operator and eigenfunctions {#laplace-operator-and-eigenfunctions .unnumbered}
The Laplace operator has eigenvalues $-(k^2+l^2)$, $k,l\in 0,1,2,\ldots,$ with eigenfunctions $\cos(k x)\cos(l y),\sin(k x)\cos(l y),\cos(kx)\sin(ly),\sin(kx)\sin(ly)$, $k,l\in 0,1,2,\ldots$. Thus the multiplicity of the first nontrivial eigenvalue -1 is 4, and the corresponding eigenspace is spanned by $\{\cos(x),\sin(x),\cos(y),\sin(y)\}$. The upper bound for $\mathbf{h}$ provided by Cheeger’s inequality is 2. Note that the zero level sets of the functions $\{\cos(x),\sin(x),\cos(y),\sin(y)\}$ are exactly the optimal disconnecting curves $\Gamma$ discussed above. The zero level set of one of these functions, $\cos(x)$ is shown in black in Figure \[torusfig\].
The rectangle {#sect:rect}
-------------
Consider the rectangle $[0,a]\times [0,b]$, with $a>b$.
### Solution to the isoperimetric problem {#solution-to-the-isoperimetric-problem-1 .unnumbered}
The problem (\[cheegerconst\]) has a unique solution $\Gamma$: $\{(a/2,y):0\le y\le b\}$, shown as a black line in Figure \[rectfig\]. The corresponding value of $\mathbf{h}$ is $2/a$.
### Laplacian operator and eigenfunctions {#laplacian-operator-and-eigenfunctions .unnumbered}
The Laplace operator with zero Neumann boundary conditions has eigenvalues $-\pi^2(k^2/a^2+l^2/b^2)$, $k,l\in 0,1,2,\ldots,$ with corresponding eigenfunctions $\cos(k\pi x/a)\cos(l\pi y/b)$, $k,l\in 0,1,2,\ldots$. Note that the Neumann boundary conditions are satisfied on the boundary of the rectangle by this set of eigenfunctions. The first nontrivial eigenvalue is $\lambda_2=-\pi^2/a^2$. In the example shown in Figure \[rectfig\], $a=3/2$, $b=1$ and the first three eigenvalues are $0,-4\pi^2/9,-\pi^2$, each of unit multiplicity, corresponding to $(k,l)=(0,0),(1,0),(0,1)$. The corresponding eigenspaces are spanned by $1,\cos(3\pi x/2),\cos(\pi y)$. Note that the zero level set of the second eigenfunction $\cos(3\pi x/2)$ is exactly the optimal disconnecting curve $\Gamma$, shown in black in Figure \[rectfig\]. The upper bound for $\mathbf{h}$ provided by Cheeger’s inequality is $2\sqrt{4\pi^2/9}=4\pi/3$.
The cylinder {#sect:cyl}
------------
Consider the flat cylinder $a(\mathbb{R}/\mathbb{Z})\times [0,b)$, which we write as $[0,a)/\sim\times [0,b]$, with $a>b$ and the vertical “edges” identified.
### Solution to the isoperimetric problem {#solution-to-the-isoperimetric-problem-2 .unnumbered}
The solution to the isoperimetric problem depends on the relative size of $a$ to $b$. If $a<2b$ then $\mathbf{h}=a$, with a unique minimising disconnecting curve $\Gamma$: $\{(x,b/2):0\le x\le a\}$. If $a>2b$ then $\mathbf{h}=2b$, with $\Gamma$ selected from an infinite family of pairs of vertical lines parameterised by $x\in[0,a)$: $\{(x,y):0\le y\le b\}\cup\{(x+a/2,y):0\le y\le b\}$.
### Laplacian operator and eigenfunctions {#laplacian-operator-and-eigenfunctions-1 .unnumbered}
The Laplace operator has eigenvalues $-\pi^2(4k^2/a^2+l^2/b^2)$, $k,l\in 0,1,2,\ldots,$ with corresponding eigenfunctions $\sin(2k\pi x/a)\cos(l\pi y/b),\cos(2k\pi x/a)\cos(l\pi y/b)$, $k,l\in 0,1,2,\ldots$. Note that we only have to enforce Neumann boundary conditions on the top and bottom horizontal boundaries of the cylinder. The leading eigenvalue is 0, and the second eigenvalue depends on the relative size of $a$ and $b$; a switch occurs at $a=2b$, matching the corresponding switch in the domain geometry. In the example shown in Figure \[cylfig\], $a=3/2$, $b=1$ and the first three eigenvalues are $0,-\pi^2,-16\pi^2/9$, with multiplicities 1, 1, and 2, corresponding to $(k,l)=(0,0),(0,1),(1,0)$. The corresponding eigenspaces are spanned by $1,\cos(\pi y),\{\sin(4\pi x/3),\cos(4\pi x/3)\}$. The upper bound for $\mathbf{h}$ provided by Cheeger’s inequality is $2\pi$. Note that the zero level set of the second eigenfunction $ \cos(\pi y)$ is exactly the optimal disconnecting curve $\Gamma$, shown in Figure \[cylfig\].
Dynamic Isoperimetry
====================
In this section we extend the concepts of the previous section to a dynamic setting, where $T$ is a $C^\infty$ diffeomorphism from $M$ onto $T(M)$, and $M$, $T(M)$ are compact, connected Riemannian manifolds of vanishing curvature. Much of the discussion in this paper is for a single iterate of a $T:M\to T(M)$, however, the extension to multiple iterates of the same map, iterates of different maps as would occur in time-dependent dynamical systems, and even a continuum of flow maps generated by a time-dependent ODE is straightforward (see Section \[sect:multistep\]). For a single iterate of $T$ we seek sets that have small boundary size relative to volume both *before* and *after* the application of the nonlinear dynamics of $T$. Thus, if $\Gamma$ is the boundary of a coherent set, one needs to minimise both $\ell_{d-1}(\Gamma)$ and $\ell_{d-1}(T\Gamma)$. To identify finite-time coherent sets, we propose the following natural dynamic minimisation problem.
\[cheegerdefn\] Define the *dynamic Cheeger constant* $\mathbf{h}^D$ by $$\label{cheegereqn2}
\mathbf{h}^D:=\inf_\Gamma \frac{\ell_{d-1}(\Gamma)+\ell_{d-1}(T(\Gamma))}{2\min\{\ell(M_1),\ell(M_2)\}}.$$ where $\Gamma$ varies over compact $(d-1)$-dimensional $C^\infty$ submanifolds of $M$ that divide $M$ into two disjoint open submanifolds $M_1, M_2$ of $M$.
In the present paper, to avoid obscuring the key constructions, we focus on volume-preserving $T$. Note that (\[cheegereqn2\]) cannot be decomposed into two static minimisation problems because $\Gamma$ is the same in both terms in the numerator of (\[cheegereqn2\]).
A Dynamic Federer-Fleming Theorem {#sect:ff}
---------------------------------
It is of theoretical interest (and for the present paper, of interest in dynamical systems applications) to connect the set-based optimisation problem (\[cheegereqn2\]) with functional optimisation problems. Two basic tools in differential geometry for doing this are the co-area formula, which connects spatial integrals of the gradient of a function with an integral over the areas of level sets of a function, and Cavalieri’s principle, which represents a function as an integral over its level sets. The Federer-Fleming theorem (Theorem \[ffthm\]) connects a set-based isoperimetric problem (the (static) Cheeger constant) in an exact way with a functional minimisation problem. We wish to formulate a dynamic equivalent of this theorem. Let $M$ be a compact, connected Riemannian manifold of dimension $d\ge 1$ with vanishing curvature, and $T:M\to T(M)$ a volume-preserving diffeomorphism. We denote by $\mathcal{P}$ the Perron-Frobenius operator of $T$, defined by $\mathcal{P}f=f\circ T^{-1}$ as $T$ is volume-preserving.
\[sobolevdefn\] Define the *dynamic Sobolev constant of $M$*, $\mathbf{s}^D(M)$ by $$\label{soboleveqn}
\mathbf{s}^D=\inf_{f\in C^\infty} \frac{\|\nabla f\|_1+\|\nabla(\P f)\|_1}{2\inf_{\alpha\in \mathbb{R}} \|f-\alpha\|_{1}}.$$
Related to the above is the alternate Sobolev constant $$\label{altsoboleveqn}
\hat{\mathbf{s}}^D=\inf_{f\in C^\infty} \frac{\|\nabla f\|_1+\|\nabla(\P f)\|_1}{2\|f-\bar{f}\|_{1}},$$ setting $\alpha$ to be the mean value of $f$, $\bar{f}=(1/\ell(M))(\int_M f\ d\ell)$ (see e.g. [@chavelisoperimetric] p163 for the static version). Clearly, $\hat{\mathbf{s}}^D\le \mathbf{s}^D$. We wish to demonstrate a dynamic analogue of the Federer-Fleming theorem. Our first main result is:
\[fflemma\] Let $M$ be a compact, connected $C^\infty$ manifold with vanishing curvature. Let $T:M\to T(M)$ be a $C^\infty$ volume-preserving diffeomorphism. Then $$\label{ffeqn}
\mathbf{s}^D=\mathbf{h}^D,$$ and further, $$\label{ffeqn22}
\mathbf{h}^D/2\le \hat{\mathbf{s}}^D\le \mathbf{s}^D=\mathbf{h}^D.$$
See appendix.
A Dynamic Cheeger Inequality {#sect:cheegerineq}
----------------------------
The (static) Cheeger inequality (Theorem \[cheegerthm\]) is an $L^2$-based result while the (static) Federer-Fleming equality (Theorem \[ffthm\]) is $L^1$-based. The advantage of $L^2$ is that one obtains a nice spectral theory for $\triangle$ from the Hilbert space structure, and crucial variational characterisations of the eigenvalues. One pays for this convenience by obtaining an inequality, rather than equality. Nevertheless, as we have seen in Sections \[sect:torus\]–\[sect:cyl\], the level sets of the Laplacian eigenfunctions carry significant information and provide good solutions to the original set-based isoperimetric problem (\[cheegerconst\]). We wish to replicate these properties for the dynamic Cheeger constant $\mathbf{h}^D$ and a dynamic version of the Laplace operator. We define the latter by $$\label{hattriangle}
\hat{\triangle}:=(\triangle+\mathcal{P}^*\triangle\mathcal{P})/2.$$ The spectral properties of this operator are developed in Section \[sect:dynlapspec\], but we say a few words here about the intuition behind this definition. Consider a function $f:M\to \mathbb{R}$ on $M$ at the initial time from which we extract level sets, as in Figures (\[torusfig\])–(\[cylfig\]). The first term in (\[hattriangle\]), $\triangle$, is the Laplacian on the domain $M$ and of obvious importance for providing information on decompositions of $M$. The second term $\mathcal{P}^*\triangle\mathcal{P}$ first pushes the function on $f$ on $M$ forward to a function $\P f$ on $T(M)$, possibly undergoing nonlinear distortion. One then applies the Laplacian to $\P f$ on $T(M)$ to obtain geometric information on $T(M)$, and finally pulls the result back to $M$ with $\P^*$, ready to be combined with the result from the first term $\triangle$. We note that in fact $\mathcal{P}^*\triangle\mathcal{P}$ is the Laplace-Beltrami operator for the pullback of the Euclidean metric on $T(M)$. Consider $\triangle_\delta:C^\infty(T(M),\mathbb{R})\circlearrowleft$ as the Laplace-Beltrami operator on the Riemannian manifold $(T(M),\delta)$, where $\delta$ denotes the Riemannian metric (in the present context, $\delta$ is the trivial Euclidean metric). Pulling $\delta$ back under $T$ we obtain the Riemannian metric $T^*\delta$ and the map $T:(M,T^*\delta)\to (T(M),\delta)$ is an isometry. One can now write $\triangle_{T^*\delta}f=(\triangle_\delta(f\circ T^{-1}))\circ T=\mathcal{P}^*\triangle_\delta\mathcal{P}$; see e.g. p27 [@chaveleigenvalues].
Our second new result is a dynamic Cheeger inequality, which highlights the importance of eigenfunctions of the operator $\hat{\triangle}$.
\[cheegerdynthm\] Let $M$ be a compact, connected $C^\infty$ manifold with vanishing curvature, and $T:M\to T(M)$ be a $C^\infty$ and volume-preserving diffeomorphism.
1. If $M$ is boundaryless, then let $\lambda_2$ be the smallest magnitude nonzero eigenvalue of $\hat{\triangle}$.
2. If $\partial M\neq\emptyset$, denote by $\mathbf{n}(x)$ the outward unit normal at $x\in \partial M$. Let $\lambda_2$ be the smallest magnitude nonzero eigenvalue for the $L^2$-eigenproblem $$\label{strongeqn0}
\hat{\triangle}u(x)=\lambda u(x),\quad x\in \mathring{M},$$ with boundary condition $$\label{strongbc0}
\nabla u(x)\cdot\left[\left(I+DT(x)^{-1}\left(DT(x)^{-1}\right)^\top\right) \mathbf{n}(x)\right]=0,\quad x\in\partial M.$$
Then $$\label{dyncheegereqn}
\mathbf{h}^{D}\le 2\sqrt{-\lambda_2}.$$
See appendix.
An intuitive explanation of the term $\left((\nabla u(x))^\top DT(x)^{-1}\right)\cdot\left(\left(DT(x)^{-1}\right)^\top\mathbf{n}(x)\right)$ in (\[strongbc0\]) is that (i) $(\nabla u(x))^\top DT(x)^{-1}=\nabla(u\circ T^{-1})(T(x))$ (gradient of the pushforward of $u$ by $T$ at $T(x)$) and (ii) $\left(DT(x)^{-1}\right)^\top\mathbf{n}(x)$ is normal to $\partial T(M)$ at $T(x)$. Thus, $\left((\nabla u(x))^\top DT(x)^{-1}\right)\cdot\left(\left(DT(x)^{-1}\right)^\top\mathbf{n}(x)\right)=0$ can be viewed as a natural pullback of a zero Neumann boundary condition on $\partial T(M)$ at $T(x)$. In terms of metrics, one has $\nabla_\delta(u\circ T^{-1})=(\nabla_{T^*\delta}(u))\circ T^{-1}$.
\[proofremark\] Using the above pullback interpretation of the boundary condition and the pullback interpretation of $\hat{\triangle}$, one can produce shorter, coordinate-free proofs of Theorems \[cheegerdynthm\] and \[dynlapthm\], instead of the coordinate-based proofs in the Appendix. Similarly, the proof of Theorem \[fflemma\] can also be easily approached from this point of view.
Multiple time-steps {#sect:multistep}
-------------------
Let us now consider a composition of several maps $T_1,\ldots,T_{n-1}$, denoting $T^{(i)}:=T_i\circ\cdots\circ T_2\circ T_1$, $i=1,\ldots,n-1$. These maps might arise, for example, as time-$\tau$ maps of a time-dependent flow. If we wish to track the evolution of a coherent set under these maps, penalising the boundary of the evolved set $T^{(i)}(\Gamma)$ after the application of each $T_i$, then we can define $$\label{hdynn}
\mathbf{h}_n^{D}:=\inf_\Gamma \frac{\frac{1}{n}\sum_{i=0}^{n-1}\ell_{d-1}(T^{(i)}\Gamma)}{\min\{\ell(M_1),\ell(M_2)\}},$$ as the natural generalisation of $\mathbf{h}^D$.
In continuous time, we consider a (possibly time-dependent) ODE $\dot{x}=F(x,t)$, where $F$ is $C^\infty$ on $M\times[0,\tau]$. The flow maps $T^{(t)}:M\to T^{(t)}(M)$ are then smooth[^2] for each $t\in[0,\tau]$. One can define $$\label{hdynt}
\mathbf{h}_{[0,\tau]}^{D}:=\inf_\Gamma \frac{\frac{1}{\tau}\int_0^\tau \ell_{d-1}(T^{(t)}\Gamma)\ dt}{\min\{\ell(M_1),\ell(M_2)\}},$$ as a time-continuous generalisation of $\mathbf{h}^D$. Analogously, setting $\mathcal{P}^{(i)}f=f\circ (T^{(i)})^{-1}$ and $\mathcal{P}^{(t)}f=f\circ (T^{(t)})^{-1}$, one can define dynamic Sobolev constants for multiple discrete time steps or over a continuous time interval: $$\label{sdynnt}
\mathbf{s}^D_n=\inf_{f\in C^\infty} \frac{\frac{1}{n}\sum_{i=0}^{n-1}\|\nabla ( \mathcal{P}^{(i)}f)\|_1}{\inf_{\alpha\in \mathbb{R}} \|f-\alpha\|_{1}},\qquad\qquad \mathbf{s}^D_{[0,\tau]}=\inf_{f\in C^\infty} \frac{\frac{1}{\tau}\int_0^\tau\|\nabla ( \mathcal{P}^{(t)}f)\|_1\ dt}{\inf_{\alpha\in \mathbb{R}} \|f-\alpha\|_{1}}.$$
\[ffcor\] Let $M$ be a compact, connected $C^\infty$ manifold with vanishing curvature, and $T^{(i)}$, $i=1,\ldots,n-1$ (resp. $T^{(t)}, t\in[0,\tau]$) be generated by a sequence of $C^\infty$ volume-preserving diffeomorphisms (resp. be smooth flow maps generated by a volume-preserving ODE $\dot{x}=F(x,t)$). Then $$\label{ffeqnn}
\mathbf{h}^D_n/2\le \hat{\mathbf{s}}^D_n\le \mathbf{s}^D_n=\mathbf{h}^D_n,$$ resp.$$\label{ffeqnt}
\mathbf{h}^D_{[0,\tau]}/2\le \hat{\mathbf{s}}^D_{[0,\tau]}\le \mathbf{s}^D_{[0,\tau]}=\mathbf{h}^D_{[0,\tau]}.$$
See appendix.
Theorem \[cheegerdynthm\] also naturally extends to multiple time steps.
\[cheegerncor\] Let $M$ be a compact, connected $C^\infty$ manifold with vanishing curvature, and $T^{(i)}$, $i=1,\ldots,n-1$ be generated by a sequence of $C^\infty$ volume-preserving diffeomorphisms. Define $$\label{bartrianglen}
\hat{\triangle}^{(n)}:=\frac{1}{n}\sum_{i=0}^{n-1}(\mathcal{P}^{(i)})^*\triangle\mathcal{P}^{(i)},$$ where $\mathcal{P}^{(i)}f=f\circ (T^{(i)})^{-1}$.
1. If $\partial M=\emptyset$, let $\lambda_2$ be the smallest magnitude nonzero eigenvalue of $\hat{\triangle}^{(n)}$.
2. If $\partial M\neq\emptyset$, denote by $\mathbf{n}(x)$ the outward unit normal at $x\in \partial M$. Let $\lambda_2$ be the smallest magnitude nonzero eigenvalue for the $L^2$ eigenproblem $$\label{strongeqnmultn}
\hat{\triangle}^{(n)}u(x)=\lambda u(x),\quad x\in \mathring{M},$$ with boundary condition $$\label{strongbcmultn}
\nabla u(x)\cdot\left[\sum_{i=0}^{n-1}DT^{(i)}(x)^{-1}\left(DT^{(i)}(x)^{-1}\right)^\top \mathbf{n}(x)\right]=0,\quad x\in\partial M.$$
Then $$\label{cheegermultistepn}
\mathbf{h}_n^{D}\le 2\sqrt{-\lambda_2}.$$
See Appendix.
\[cheegertcor\] Let $M$ be a compact, connected $C^\infty$ manifold with vanishing curvature, and $T^{(t)}, t\in[0,\tau]$ be smooth flow maps. Define $$\label{bartrianglet}
\hat{\triangle}^{(\tau)}:=\frac{1}{\tau}\int_0^\tau(\mathcal{P}^{(t)})^*\triangle\mathcal{P}^{(t)}\ dt,$$ where $\mathcal{P}^{(t)}f=f\circ (T^{(t)})^{-1}$.
1. If $\partial M=\emptyset$, let $\lambda_2$ be the smallest magnitude nonzero eigenvalue of $\hat{\triangle}^{(\tau)}$.
2. If $\partial M\neq\emptyset$, denote by $\mathbf{n}(x)$ the outward unit normal at $x\in \partial M$. Let $\lambda_2$ be the smallest magnitude nonzero eigenvalue for the $L^2$ eigenproblem $$\label{strongeqnmultt}
\hat{\triangle}^{(\tau)}u(x)=\lambda u(x),\quad x\in \mathring{M},$$ with boundary condition $$\label{strongbcmultt}
\nabla u(x)\cdot\left[\int_0^\tau DT^{(t)}(x)^{-1}\left(DT^{(t)}(x)^{-1}\right)^\top \mathbf{n}(x)\ dt\right]=0,\quad x\in\partial M.$$
Then $$\label{cheegermultistept}
\mathbf{h}_\tau^{D}\le 2\sqrt{-\lambda_2}.$$
See Appendix.
If one does not wish to track the length of the evolved $\Gamma$ except at the initial and final time, one would instead use (\[hattriangle\]) with $\mathcal{P}=\mathcal{P}^{(n-1)}$ or $\mathcal{P}=\mathcal{P}^{(\tau)}$.
Spectral properties of the dynamic Laplacian {#sect:dynlapspec}
============================================
The following result summarises important properties of the operator $\hat{\triangle}=(\triangle+\P^*\triangle \mathcal{P})/2$.
\[dynlapthm\] Let $M$ be a compact, connected $C^\infty$ manifold with vanishing curvature, and $T:M\to T(M)$ be a $C^\infty$, volume-preserving diffeomorphism.
- If $M$ is boundaryless, let $\lambda, u$ denote solutions to the $L^2$ eigenproblem $(1/2)(\triangle+\mathcal{P}^*\triangle\mathcal{P})u=\lambda u$ on $M$.
- If $\partial M\neq\emptyset$, denote by $\mathbf{n}(x)$ the outward unit normal at $x\in \partial M$. Let $\lambda, u$ denote solutions to the $L^2$-eigenproblem $$\label{strongeqn}
(1/2)(\triangle+\P^*\triangle\P)u(x)=\lambda u(x),\quad x\in \mathring{M},$$ with boundary condition $$\label{strongbc}
\nabla u(x)\cdot\left[\mathbf{n}(x)+DT(x)^{-1}\left(DT(x)^{-1}\right)^\top \mathbf{n}(x)\right]=0,\quad x\in\partial M.$$
The solutions $\lambda, u$ satisfy the following properties.
1. The eigenvalues form a decreasing sequence $0=\lambda_1> \lambda_2>\cdots$ with $\lambda_n\to-\infty$.
2. The corresponding eigenfunctions $u_1,u_2,\ldots$ are $C^\infty$ on $M$ and eigenfunctions corresponding to distinct eigenvalues are pairwise orthogonal in $L^2$.
3. One has the variational characterisation of eigenvalues: if $u_1,u_2,\ldots$ are arranged to be orthonormal, denoting $X_k=\span\{u_1,u_2,\ldots,u_k\}$ $$\label{variational}
\lambda_k=-\inf_{u\in X,\langle u,u_i\rangle=0, i=1,\ldots,k-1}\frac{\int_M |\nabla u|^2\ d\ell+\int_{T(M)}|\nabla(\P u)|^2\ d\ell}{2\int_M u^2\ d\ell},$$ with the infimum achieved only when $u=u_k$.
Our main focus is the eigenvalue $\lambda_2$ and the corresponding eigenfunction $u_2$. We will see that $u_1\equiv 1$ and therefore that $\int_M u_2\ d\ell=0$ by $L^2$-orthogonality to $u_1$. Appendix \[sec:weakexist\] contains the proofs of items 1 and 3 of Theorem \[dynlapthm\] and Appendix \[sec:ellipticity\] contains the proofs of item 2 and the boundary conditions.
\[spectrummultistep\] Using identical arguments, one can also show a multiple time step version of Theorem \[dynlapthm\], where $(1/2)(\triangle+\P^*\triangle\P)$ is replaced with either (\[bartrianglen\]) or (\[bartrianglet\]), and the boundary condition (\[strongbc\]) is replaced with either (\[strongbcmultn\]) or (\[strongbcmultt\]).
Objectivity
-----------
We demonstrate that the operator $\hat{\triangle}$ behaves in a very predictable way when the phase space is observed in a time-dependent rotating and translating frame. In particular, we show that the method of extracting coherent sets from eigenvectors of $\hat{\triangle}$ (described in Section \[sect:numerics\]) is *objective* or *frame-invariant*, meaning that the method produces the same features when subjected to time-dependent “proper orthogonal + translational” transformations; see [@truesdellnoll].
In continuous time, to test for objectivity, one makes a time-dependent coordinate change $x\mapsto Q(t)x+b(t)$ where $Q(t)$ is a proper othogonal linear transformation and $b(t)$ is a translation vector, for $t\in[t_0,t_1]$. The discrete time analogue is to imagine we begin in the frame given by $\Phi_{t_0}(M)$, where $\Phi_{t_0}(x)=Q(t_0)x+b(t_0)$, and end in the frame $\Phi_{t_1}(M)$, where $\Phi_{t_1}(x)=Q(t_1)x+b(t_1)$. We are concerned with the deterministic transformation $\dot{T}:\Phi_{t_0}(M)\to \Phi_{t_1}(M)$, which is given by $\dot{T}=\Phi_{t_1}\circ T\circ \Phi_{t_0}^{-1}$. This change of frames is summarised in the commutative diagram below. $$\begin{CD}
M @>T>> T(M)\\
@VV\Phi_{t_0}V @VV\Phi_{t_1}V\\
\Phi_{t_0}(M) @>\dot{T}>> \Phi_{t_0}\circ T(M)
\end{CD}$$ Corresponding to $\dot{T}$ is the operator $\dot{\hat{\triangle}}=\triangle+\P^*_{\dot{T}}\triangle \P_{\dot{T}}$, where $\P_{\dot{T}}=\P_{\Phi_{t_1}}\circ \P\circ \P_{\Phi_{t_0}}^{-1}$. If we were observing the dynamics in the frames given by $\Phi_{t_0}$ and $\Phi_{t_1}$ we would compute eigenfunctions of $\dot{\hat{\triangle}}$.
\[objthm\] The operator $\hat{\triangle}$ in the original frame and the operator $\dot{\hat{\triangle}}$ in the transformed frame satisfy the commutative diagram: $$\begin{CD}
L^2(M) @>\hat{\triangle}>> L^2(M)\\
@VV\P_{\Phi_{t_0}}V @VV\P_{\Phi_{t_0}}V\\
L^2(\Phi_{t_0}(M))@>\dot{\hat{\triangle}}>>L^2(\Phi_{t_1}(M))
\end{CD}$$ Consequently, if $f$ solves $$\label{origevalprob}
\hat{\triangle}f=\lambda f\quad\mbox{ on $\mathring{M}$},$$ then $$\label{transfevalprob}
\dot{\hat{\triangle}}(\P_{\Phi_{t_0}}f)=\lambda (\P_{\Phi_{t_0}}f)\quad\mbox{ on $\Phi_{t_0}(\mathring{M})$}.$$ Furthermore, if $\partial M\neq \emptyset$, then if $$\label{strongorigbc}
\nabla f(x)\cdot\left[\mathbf{n}(x)+DT(x)^{-1}\left(DT(x)^{-1}\right)^\top\mathbf{n}(x)\right]=0,\quad x\in\partial M,$$ one has $$\label{strongtransfbc}
\nabla (\P_{\Phi_{t_0}}f)(x)\cdot\left[\dot{\mathbf{n}}(x)+D\dot{T}(x)^{-1}\left(D\dot{T}(x)^{-1}\right)^\top\dot{\mathbf{n}}(x)\right]=0,\quad x\in\partial(\Phi_{t_0}(M)),$$ where $\dot{\mathbf{n}}(x)=Q(t_0)\mathbf{n}(\Phi_{t_0}^{-1}x)$.
It follows from Theorem \[objthm\] that the coherent sets extracted on $M$ from e.g. level sets of the eigenfunctions of $\dot{\hat{\triangle}}$ will be transformed versions (under $\Phi_{t_0}$) of those extracted from $\hat{\triangle}$, as required for objectivity.
Zero-diffusion limit of an analytic diffusion-based framework {#sect:zerolimit}
=============================================================
The paper [@F13] introduced an analytic methodology for finding finite-time coherent sets, formalising prior numerical work [@FSM10]. This methodology was based around smoothings of $\P$. In [@F13], one defined smoothing operators $\mathcal{D}_{M,\epsilon}:L^2(M)\to \mathcal{H}_{1/2}(M_\epsilon)$, $\mathcal{D}_{T(M_\epsilon),\epsilon}:L^2(T(M_\epsilon))\to \mathcal{H}_{1/2}(T(M)_\epsilon)$ where $\mathcal{H}_{1/2}(M_\epsilon), \mathcal{H}_{1/2}(T(M)_\epsilon)$ denote Hölder functions with exponent 1/2 on $\epsilon$-neighbourhoods of $M$ and $T(M_\epsilon)$, respectively. The operators considered in [@F13] were $\mathcal{D}_{M,\epsilon} f(y)=\int_M \alpha_\epsilon(x-y)f(x)\ d\ell(x)$, $y\in M_\epsilon$ and $\mathcal{D}_{T(M_\epsilon),\epsilon} f(y)=\int_{T(M_\epsilon)} \alpha_\epsilon(x-y)f(x)\ d\ell(x)$, $y\in T(M)_\epsilon$, where $\alpha_\epsilon(x)=\mathbf{1}_{B_\epsilon(0)}/\ell(B_\epsilon(0))$, corresponds to smoothing on a local $\epsilon$-ball. In [@F13], the operator $\mathcal{L}_\epsilon:L^2(M)\to L^2(T(M)_\epsilon)$, $\mathcal{L}_\epsilon=\mathcal{D}_{T(M_\epsilon),\epsilon}\P\mathcal{D}_{M,\epsilon}$ was introduced[^3] and used to identify coherent sets for $T$ in phase space $M$. The reason for the diffusion operators $\mathcal{D}_\epsilon$ were two-fold.
1. Firstly, as $T$ is often invertible (e.g. the time-$t$ map of some smooth flow), subsets of $M$ are simply deformed by $T$, they do not “disperse”, and one could argue that every set is “coherent” in the sense that it is non-dispersive. Let us consider the action of $\mathcal{L}_\epsilon$ on $\mathbf{1}_{A}$, where the latter represents a subset $A\subset M$ by its characteristic function; we think of $\mathbf{1}_{A}$ as a uniform mass distribution on $A$. Applying $\mathcal{L}_\epsilon$, we first have $\mathcal{D}_{M,\epsilon}$ acting on $\mathbf{1}_{A}$, which removes from $A$ some mass within distance $\epsilon$ of the boundary of $A$. The resulting function is then transformed dynamically by $\P$, and will be supported on an $\epsilon$-neighbourhood of $T(A)$. Finally, we apply $\mathcal{D}_{T(M_\epsilon)}$ again, so that some mass within a distance $\epsilon$ of the boundary of the support of $\P\mathcal{D}_{M,\epsilon}\mathbf{1}_A$ is ejected from this support. These ideas are quantified in the proof of Lemma 6 [@F13]. In this way, the boundary size of both $A$ and $T(A)$ are penalised because the amount of mass ejected by the operators $\mathcal{D}_{M,\epsilon},\mathcal{D}_{T(M_\epsilon),\epsilon}$ is proportional to the boundary sizes.
2. Secondly, in order to find a set $A$ with minimal combined boundary sizes for $A$ and $T(A)$, [@F13] used minimisation properties of the singular vectors of $\mathcal{L}_\epsilon$; in particular, the sets $A$ and $T(A)$ were estimated from the left/right singular vectors corresponding to the second largest singular value of $\mathcal{L}_\epsilon$ (the leading singular value is always 1 by construction). To use this variational machinery $\mathcal{L}_\epsilon$ needs to be compact, and it was shown in [@F13] that $\mathcal{D}_{M,\epsilon}, \mathcal{D}_{T(M_\epsilon)}$ also played the technical role of ensuring compactness of $\mathcal{L}_\epsilon$ acting on $L^2$ functions.
The singular vector of $\mathcal{L}_\epsilon$ that corresponds to the initial time (prior to application of $T$) is an eigenvector of $\mathcal{A}_\epsilon:=\mathcal{L}_\epsilon^*\mathcal{L}_\epsilon$; one pushes forward and then pulls back. Without any diffusion operators, this would read $\mathcal{A}_0=\P^*\P$; deterministically pushing forward and deterministically pulling back. Because $T$ is volume-preserving and invertible, $\P f=f\circ T^{-1}$ and $\P^*=f\circ T$. Thus $\mathcal{A}_0$ is the identity operator, and one lacks compactness and a “second” eigenvalue. Without diffusion, there is no distinguished coherent set, all sets are equally coherent as they are merely distorted, not dispersed, by the deterministic dynamics over the finite time duration encoded in $T$.
However, one can ask about higher order terms when $\epsilon$ is close to zero. We show that with the right scaling in $\epsilon$, one can make sense of an expression like $$\label{difflimit}
\mathcal{B}f(x):=\lim_{\epsilon\to 0}\frac{(\mathcal{L}_\epsilon^*\mathcal{L}_\epsilon-I)f(x)}{\epsilon^\beta},$$ with $\mathcal{B}$ capturing the essential effects of tiny $\epsilon$-diffusion *without explicitly including* that diffusion.
We slightly modify and generalise the diffusion operators from [@F13]. Let $q:M\to\mathbb{R}^+$ be a nonnegative density with compact support, with mean the origin, and with covariance matrix $c\cdot I$, where $I$ is the $d\times d$ identity matrix. We scale $q$ to form $q_\epsilon(x)=q(x/\epsilon)/\epsilon^d$; $q_\epsilon$ will play the role of the previous $\alpha_\epsilon$, and obviously $q(x)=\mathbf{1}_{B_1(0)}/\ell(B_1(0))$ is one example of a density satisfying the above conditions. We redefine $\mathcal{D}_{M,\epsilon}f(x)=\int_M q_\epsilon(x-y)f(y)\ d\ell(y)$, $x\in \mathring{M}$, and $\mathcal{D}_{T(M),\epsilon}f(x)=\int_{T(M)} q_\epsilon(x-y)f(y)\ d\ell(y)$, $x\in \mathring{T(M)}$, where $\epsilon=\epsilon(x)$ is sufficiently small that both operators preserve integrals (i.e. $\int_M f\ d\ell=\int_M \mathcal{D}_{M,\epsilon}f\ d\ell$ and $\int_{T(M)} f\ d\ell=\int_{T(M)} \mathcal{D}_{T(M),\epsilon}f\ d\ell$). In the sequel we use the definition $\mathcal{L}_\epsilon=\mathcal{D}_{T(M),\epsilon}\P\mathcal{D}_{M,\epsilon}$. With the additional assumptions $\int_M q_\epsilon(x-y)^2\ d\ell(y)d\ell(x), \int_{T(M)} q_\epsilon(x-y)^2\ d\ell(y)d\ell(x)<\infty$, one has $\mathcal{L}_\epsilon:L^2(M)\to L^2(T(M))$ is compact as required in [@F13]. The following theorem shows that one can in fact take the scaling limit (\[difflimit\]) with $\beta=2$ and that $\mathcal{B}$ is a scalar multiple (the variance of the diffusion) of $\hat{\triangle}$.
\[analcvgce\] Let $M$ be a connected, compact Riemannian manifold of vanishing curvature, $f:M \to\mathbb{R}$ be $C^3$, and $T:M\to T(M)$ be $C^3$ and volume-preserving. Let $q:M\to\mathbb{R}^+$ be a nonnegative density with compact support, with mean the origin, and covariance matrix $c\cdot I$, where $I$ is the $d\times d$ identity matrix, and let $\mathcal{L}_\epsilon=\mathcal{D}_{T(M),\epsilon}\P\mathcal{D}_{M,\epsilon}$ be defined as above. One has $$\label{diffform1thm}\lim_{\epsilon\to 0} \frac{(\mathcal{L}_\epsilon^*\mathcal{L}_\epsilon-I)f(x)}{\epsilon^2}=c\cdot(\triangle+\P^*\triangle\P)f(x),$$ for each $x\in \mathring{M}$.
The proof of Theorem \[analcvgce\] is in Appendix \[sec:analcvgce\]. The appearance of the Laplace operator is due to the fact that $\mathcal{D}_{M,\epsilon}f(x)\approx f(x)+ (c\epsilon^2/2)\triangle f(x)$ for small $\epsilon$. The symmetry conditions on $q$ in Theorem \[analcvgce\] (which are physically desirable as they model isotropic diffusion) cause the first order term in $\epsilon$ to vanish.
\[unifexample\] If $q(x)=\mathbf{1}_{B_1(0)}/\ell(B_1(0))$ (uniform diffusion on a unit ball), then $c=1/3, 1/4, 1/5$ in dimensions $d=1, 2, 3$, respectively.
Theorem \[analcvgce\] provides a theoretical link between the diffusion-based method [@F13] and the diffusion-free constructions based on the Laplace operator in the present paper. The latter have very strong connections with geometry, evidenced by Theorems \[fflemma\] and \[cheegerdynthm\], and further reinforce the geometric intuition of [@F13]. In numerical computations, if the dynamical system is deterministic and the dynamics and the domain are smooth, the present construction may be advantageous because the spectrum of $\hat{\triangle}$ is well-separated, while the second-largest eigenvalue of $\mathcal{L}_\epsilon^*\mathcal{L}_\epsilon$ is likely to be separated from 1 by order $\epsilon^2$. If the dynamical system or the domain lacks smoothness, or if the dynamics has nontrivial diffusion from a model, both of which are not uncommon in many real-world applications, then approach of [@F13] may be more appropriate. The nontrivial diffusion from the model will in this case produce a larger spectral gap.
Numerical experiments {#sect:numerics}
=====================
In this section, we propose a method for finding coherent sets with low Cheeger ratios $\mathbf{h}^D(\Gamma):=(\ell_{d-1}(\Gamma)+\ell_{d-1}(T\Gamma))/\min\{\ell_d(M_1),\ell_d(M_2)\}$, where $\Gamma$ disconnects $M$ into $M_1,M_2$. We use the level sets of the first nontrivial eigenfunction of the dynamic Laplacian ${\hat{\triangle}}$, in analogy to the level sets of the first nontrivial eigenfunction of the Laplacian $\triangle$ in the static case described in Section \[staticsection\]. Our goal here is to demonstrate the efficacy of this approach, rather than to find the most accurate or efficient numerical implementation, which will be treated in a forthcoming study.
To numerically estimate the Perron-Frobenius operator $\mathcal{P}$ we use Ulam’s method [@ulam]. For simplicity, we describe here the case of $T(M)=M$; the construction for $T(M)\neq M$ is completely analogous and can be found in [@FSM10; @FPG14]. We partition $M$ into a grid of $n$ small boxes $\{B_1,\ldots,B_n\}$ and compute a matrix $P$ of conditional transition probabilities between boxes under the action of $T$. Using a uniform intra-box grid of $Q$ points $z_{i,1},\ldots,z_{i,Q}\in B_i$, one computes $P_{ij}=\#\{z_{i,q}\in B_i: T(z_{i,q})\in B_j\}/\#\{z_{i,q}\in B_i\}$. The matrix $P$ is a row-stochastic matrix, where the $(i,j)^{\rm th}$ entry estimates the conditional probability of a randomly chosen point in $B_i$ entering $B_j$ under the application of $T$. The connection with $\mathcal{P}$ is as follows. Denote by $\pi_n:L^1(M)\to\sp\{\mathbf{1}_{B_1},\ldots,\mathbf{1}_{B_n}\}$ the projection onto characteristic functions on grid sets. One has $[\pi_n\mathcal{P}f]=\tilde{P}[\pi_nf]$, where $\tilde{P}$ is the transpose of $P$ and $[f]$ denotes the vector formed from the $n$ values taken by an $f\in \sp\{\mathbf{1}_{B_1},\ldots,\mathbf{1}_{B_n}\}$.
The Laplace operator $\triangle$ in our two-dimensional examples is approximated using finite-difference on a five-point stencil, calculated at the centre points of the grid boxes $\{B_1,\ldots,B_n\}$. We treat the cases where $M$ has boundary rather crudely, simply applying zero Neumann boundary conditions via a symmetric reflection in the finite-difference scheme, without directly enforcing (\[strongbc\]). For example, given an $N\times N'$ grid covering a rectangle, denote $f_{i,j}$ to be the value of $f$ at grid position $(x_i,y_j)$. At the right-hand boundary $f_{N,j}$, we replace the fictional extension $f_{N+1,j}$ in the usual five-point stencil $f_{N+1,j}+f_{N-1,j}+f_{N,j+1}+f_{N,j-1}-4f_{N,j}$ with a symmetric extension $f_{N+1,j}=f_{N-1,j}$ to obtain $2f_{N-1,j}+f_{N,j+1}+f_{N,j-1}-4f_{N,j}$. The resulting matrix is denoted $\triangle_n$. We note that the matrices $\tilde{P}$ and $\triangle_n$ are sparse and consequently $\hat{\triangle}_n=\triangle_n+\tilde{P}^\top\triangle_n\tilde{P}$ is also sparse. The boxes $\{B_1,\ldots,B_n\}$ and matrix $P$ were constructed in Matlab using the GAIO software [@DFJ01]. The level sets of the eigenfunctions of the approximation of $\hat{\triangle}_n$ are extracted automatically using Matlab’s `contour` function, with the default settings.
The algorithm we use in the following two-dimensional case studies is described below.
\[mainalg\]
1. Form the matrix $\tilde{P}$ and the discrete Laplacian $\triangle_n$ as described above, and combine to create $\hat{\triangle}_n$.
2. Calculate eigenvalues $\lambda_1>\lambda_2>\cdots$ and eigenvectors $u_1, u_2,\ldots$ of $\hat{\triangle}_n$.
3. Iteratively scan over values of $u_2$ from $\min_i u_{2,i}$ to $\max_i u_{2,i}$. For each value, extract a level curve $\Gamma$ in $M$ using Matlab’s `contour` function (this function returns a collection of points representing corners of a polygonal curve). To compute $T\Gamma$,
1. Either: map the points representing $\Gamma$ directly with $T$,
2. Or: compute $\tilde{P}u_2$ and extract $T\Gamma$ using Matlab’s `contour` function with the same level set value as for $\Gamma$.
4. Optimise $\mathbf{h}^D(\Gamma)$ by running over all curves $\Gamma$ formed from level sets of $u_2$ in Step 3. The length of $\Gamma$ and $T\Gamma$ are computed as the lengths of the polygonal curves comprising them. Report the $\Gamma$ and $T\Gamma$ that yield the lowest value of $\mathbf{h}^D(\Gamma)$.
Linear shear on a cylinder
--------------------------
Our first example is a linear shear on a cylinder $M=[0,4)/\sim\times [0,1]$, where the $x$-coordinate is periodic. The map $T:M\circlearrowleft$ is the horizontal shear $T(x,y)=(x+y,y)$. We begin by exploring some naive guesses for an optimal $\Gamma$. Choosing $\Gamma$ to be $\{(x,1/2):0\le x<4\}$ separates the cylinder into upper and lower halves, and such a $\Gamma$ is preserved by $T$; the length of $\Gamma$ and $T(\Gamma)$ are both relatively long at 4 units each, and $\mathbf{h}^D(\Gamma)=(4+4)/(2\times 2)=2$. On the other hand, choosing $\Gamma=\{(x,y):0\le y\le 1\}\cup \{(x+2,y):0\le y\le 1\}$ separates the cylinder into two rectangles. In this case, the length of $\Gamma$ is 2, while the length of $T(\Gamma)$ is $2\sqrt{2}$; $\mathbf{h}^D(\Gamma)=(2+2\sqrt(2))/(2\times 2)=(1+\sqrt{2})/2$, an improvement over our previous guess.
The numerical computations are carried out using a $256\times 64$ grid of $2^{14}$ square boxes and within each box, $Q=1600$ test points are used to estimate the entries of $P$. The eigenvalues of $\hat{\triangle}_n$ are $-0.0271, -3.0865, -3.1368, -10.2103, -12.3406, -12.3769,\ldots$. The first eigenvalue is not exactly zero because the constant vector is not mapped exactly to a constant vector by $\tilde{P}$ due to finite point sampling in its construction. We use the eigenvector corresponding to $\lambda_2=-3.0865$ to estimate coherent sets. The results are shown in Figures \[shearvecs\] and \[shearcs\].
![Shear map: The second eigenvector and its image under $\mathcal{P}$ are shown, in addition to the optimised level set at $-1.0457\times 10^{-4}$.[]{data-label="shearvecs"}](shearvecs.png "fig:"){width="20cm"}\
![Shear map: The extracted coherent sets at the optimised level set at $- 1.0457\times 10^{-4}$.[]{data-label="shearcs"}](shearcs.png "fig:"){width="20cm"}\
In this simple example, one can calculate exactly that $u_2(x,y)=\sin((x+y/2)\pi/2)$ is an eigenfunction of $\hat{\triangle}$, with eigenvalue $5\pi^2/16$ (multiplicity 2). One may construct a one-parameter family of optimal coherent sets by sliding the sets in Figure \[shearcs\] (left) sideways, with corresponding movement of the sets in Figure \[shearcs\] (right). The boundaries of the members of this family are of the form $\Gamma=\{(x-y/2,y):0\le y\le 1\}\cup \{(x-y/2+2,y):0\le y\le 1\}$ (parameterised by $x\in [0,4)$). As the lengths of both $\Gamma$ and $T(\Gamma)$ are both $\sqrt{5}$, we can compute exactly the Cheeger constant using $\ell_d(M_1)=\ell_d(M_2)=2$ to obtain $\mathbf{h}^D(\Gamma)=(\sqrt{5}+\sqrt{5})/(2\times 2)=\sqrt{5}/2$. Thus, the second eigenfunction of $\hat{\triangle}$ is “balancing” the boundary lengths between the initial and final times in order to optimise the sum of these lengths. Note that $\sqrt{5}/2$ further improves over the Cheeger value $(1+\sqrt{2})/2$ of our second naive solution of above.
We note that the boundary condition (\[strongbc\]) is automatically satisfied by $u_2(x,y)=\sin((x+y/2)\pi/2)$. For example, the outward normal vector on the lower boundary of $M$ and $T(M)$ is $\mathbf{n}(x)\equiv [0, -1]^\top$, and $DT^{-1}(x,y)\equiv \begin{pmatrix}1 & -1 \\ 0 & 1 \end{pmatrix}$, so the condition (\[strongbc\]) is that $\nabla u_2(x,0)\cdot ([0, -1]+[1, -1])^\top=0$, which is clearly satisfied. The numerically computed eigenfunction in Figure \[shearvecs\] also appears to satisfy this condition, even with the relatively crude numerical scheme we have employed.
In comparison with the numerics, Algorithm \[mainalg\] produces $\ell_d(M_1)=\ell_d(M_2)=2$ (to 4 significant figures), while the value for $\ell_{d-1}(\Gamma)+\ell_{d-1}(T\Gamma)$ is around $1\%$ too low because the Matlab’s contour function does not extend all the way to the cylinder boundary because of the box discretisation. The bound for the Cheeger constant from Theorem \[cheegerdynthm\] is 3.5137, a consistent upper bound for the exact value of $\mathbf{h}^D$.
The remaining eigenfunctions of $\hat{\triangle}$ provide good independent solutions to the dynamic boundary minimising problem. By Theorem \[dynlapthm\], the eigenfunctions of $\hat{\triangle}$ corresponding to distinct eigenvalues are mutually orthogonal. Thus if we extract coherent sets from different eigenfunctions using the level set approach, we obtain solutions that are “independent”, in the sense that one is not a small perturbation of another. In this example, one can exactly compute that $u(x,y)=\cos(\pi y)$ is an eigenfunction with eigenvalue $\pi^2$ (unit multiplicity), and $u(x,y)=\sin((x+y/2)\pi)$ with eigenvalue $5\pi^2/4$ (multiplicity 2). The eigenvalues $5\pi^2/16, \pi^2,$ and $5\pi^2/4$ are the eigenvalues numbered two to six numerically computed (approximately) above. The numerically computed eigenfunctions are shown in Figure \[shear3efuncs\], and it is clear that zero level sets of these eigenfunctions provide a ranking of good independent solutions of decreasing quality (increasing total boundary length).
![Shear map: The second, fourth, and fifth eigenvectors of $\hat{\triangle}$ (top to bottom).[]{data-label="shear3efuncs"}](shear3efuncs.png "fig:"){width="12cm"}\
The standard map on the torus
-----------------------------
Our second example is nonlinear dynamics on a flat boundaryless manifold: the so-called “standard map” $T:\mathbb{T}^2\circlearrowleft$ on the 2-torus is given by $T(x,y)=(x+y,y+8\sin(x+y))\pmod{2\pi}$. We begin by testing a naive guess for the optimal $\Gamma$, namely one of the continuum of solutions to the static isoperimetric problem illustrated in Figure \[torusfig\]): $\Gamma=\{(\{\pi/2\}\times [0,2\pi))\cup(\{3\pi/2\}\times [0,2\pi))$. Figure \[fig:standardbenchmark\] illustrates the action of $T$ on the partition defined by $\Gamma$; while the length of $\Gamma$ is short, the nonlinear action of $T$ rapidly lengthens the boundary, and the length of $T(\Gamma)$ is much greater.
![Standard map: The black set (left) arises as one of a continuum of solutions to the static isoperimetric problem (see Figure \[torusfig\]). Its image (right) has a much longer boundary and consequently a high $\mathbf{h}^D$ value.[]{data-label="fig:standardbenchmark"}](standard_benchmark_small2.png "fig:"){width="15cm"}\
To find the optimal $\Gamma$, numerical computations are carried out using a $128\times 128$ grid of $2^{14}$ boxes and within each box, $Q=1600$ test points are used to estimate the entries of $P$. The eigenvalues of $\hat{\triangle}_n$ are $-0.1487,-1.6466,-1.6498,-6.0875,-6.0939,\ldots$. The first eigenvalue is not exactly zero because the constant vector is not mapped exactly to a constant vector by $\tilde{P}$ due to finite point sampling in its construction. We use the eigenvector corresponding to $\lambda_2=-1.6466$ to estimate coherent sets. The results are shown in Figures \[standardvecs\] and \[standardcs\].
![Standard map: The second eigenvector and its image under $\mathcal{P}$ are shown, in addition to the optimised level set at $-2.4741\times 10^{-4}$.[]{data-label="standardvecs"}](standard_vecs.png "fig:"){width="15cm"}\
![Standard map: The extracted coherent sets at the optimised level set at $-2.4741\times 10^{-4}$.[]{data-label="standardcs"}](standard_coherentsets.png "fig:"){width="15cm"}\
It is clear that the fact that the standard map creates affine dynamics in certain directions is being exploited by the operator $\hat{\triangle}$ in order to find boundaries that are initially small and remain small under one iterate of $T$ (in fact, the boundary length is reduced under $T$). One may construct a one-parameter family of optimal coherent sets by sliding the sets in Figure \[standardcs\](b) sideways, with corresponding movement of the sets in Figure \[standardcs\](a). The second eigenvalue of $\hat{\triangle}$ is therefore probably of multiplicity 2, and this is borne out by the closeness of the computed values for $\lambda_2$ and $\lambda_3$.
In this case we can compute exactly the Cheeger constant because $\ell_d(M_1)=\ell_d(M_2)=(2\pi)^2/2$ and $\ell_{d-1}(\Gamma)=4\sqrt{2}\pi$ and $\ell_{d-1}(T\Gamma)=4\pi$. Thus $\mathbf{h}^D=(1+\sqrt{2})/\pi\approx 0.7685$. In comparison with the numerics, one obtains $\ell_d(M_1)=\ell_d(M_2)=(2\pi)^2/2$ (to 4 significant figures), while the value for $\ell_{d-1}(\Gamma)+\ell_{d-1}(T\Gamma)$ is around $1\%$ too low because the Matlab’s contour function does not extend all the way to the torus boundary. Bounds for the Cheeger constant from (\[soboleveqn\]) and Theorem \[cheegerdynthm\] are 1.2278 and 2.5664, respectively, both consistent upper bounds for the exact value of $\mathbf{h}^D$.
Transitory flow on the square
-----------------------------
Our third example is a nonlinear time-dependent flow on the unit square introduced in [@meissmosovsky], defined by $\dot{x}=-\partial\Psi/\partial y, \dot{y}=-\partial\Psi/\partial x,$ where $\Psi$ is the time-dependent stream function $\Psi(x,y,t)=(1-s(t))\sin(2\pi x)\sin(\pi y)+s(t)\sin(\pi x)\sin(2\pi y)$ and $s(t)=t^2(3-2t), 0\le t\le 1$. The flow is computed from $t=0$ to $t=1$. At time $t=0$, the instantaneous vector field comprises two separate rotating “gyres” on the left and right halves of the square. As $t$ increases from 0 to 1, the instantaneous vector field rotates 90 degrees to finally arrive at two rotating gyres in the upper and lower halves of the square.
The numerical computations are carried out using a $128\times 128$ grid of $2^{14}$ boxes and within each box, $Q=1600$ test points are used to estimate the entries of $P$. The eigenvalues of $\hat{\triangle}_n$ are $-39.9269$, $-87.1430$, $-155.7652$, $-352.8106$, $-430.3017$, $-465.4415,\ldots$ The first eigenvalue is again not exactly zero, because the constant vector is not mapped exactly to a constant vector by $\tilde{P}$ due to finite point sampling in its construction. We use the eigenvector corresponding to $\lambda_2=-87.1430$ to estimate coherent sets. The results are shown in Figures \[meissvecs\] and \[meisscs\].
![Transitory flow: The second eigenvector and its image under $\mathcal{P}$ are shown, in addition to the optimised level set at $-6.4417\times 10^{-4}$.[]{data-label="meissvecs"}](meiss_vecs.png "fig:"){width="15cm"}\
![Transitory flow: The extracted coherent sets at the optimised level set at $-6.4417\times 10^{-4}$.[]{data-label="meisscs"}](meiss_coherentsets.png "fig:"){width="15cm"}\
From the numerics, one obtains $\ell_d(M_1)=0.3091$, $\ell_{d-1}(\Gamma)=2.1606$, $\ell_{d-1}(T\Gamma)=2.9557$ and the value for $\mathbf{h}^D(\Gamma)\approx(2.1606+2.9557)/(2\times 0.3091)=8.2749$. Bounds for the Cheeger constant from (\[soboleveqn\]) and Theorem \[cheegerdynthm\] are 10.0533 and 18.6701, respectively, both consistent upper bounds. We compare these results with a “naive” solution, where one selects $\Gamma'$ to be the vertical separatrix that separates the two rotating elements in the instantaneous vector field at $t=0$; see Figure \[meissvi\] (left). This choice of $\Gamma'$ is one of two solutions to the static isoperimetric problem on the unit square, and corresponds to a static Cheeger value of $\mathbf{h}(\Gamma')=1/(1/2)=2.$ The image of $\Gamma'$ under $T$ is shown in Figure \[meissvi\] (right).
![Transitory flow: A vertical separatrix and its image from $t=0$ to $t=1$.[]{data-label="meissvi"}](meiss_verticalimage.png "fig:"){width="15cm"}\
While the length of $\Gamma'$ is only 1, the length of $T\Gamma'$ is much greater (approximately 8.3057), leading to a Cheeger value of $\mathbf{h}^D(\Gamma')\approx(1+8.3057)/(2\times 1/2)=9.3057$, larger than the value of $\mathbf{h}^D(\Gamma)=8.2749$ corresponding to the solution shown in Figures \[meissvecs\] and \[meisscs\]. We see that the curve $\Gamma$ in Figure \[meissvecs\] trades off length at $t=0$ in order to have a relatively short length also at time $t=1$, in contrast to $\Gamma'$.
Finally, Figure \[meisszooms\] shows fine detail of the curves $T(\Gamma)$; the pixellation visible is the underlying grid, which controls the resolution of the boundary curves. There is some shearing at the two locations shown. This is responsible for most of the increase in $\ell_{d-1}(T(\Gamma))$ from $\ell_{d-1}(\Gamma)$.
![Transitory flow: Zooms of the boundary at time $t=1$.[]{data-label="meisszooms"}](meiss_zooms.png "fig:"){width="15cm"}\
While the shearing is not tiny, particularly in the left-hand figure, given the limited resolution and the fact that most of the boundary is shear-free, our selected coherent sets do perform well in terms of reducing boundary length for both the initial set and its image. Moreover, if one considers applying diffusion at the scale of the box diameters, the “effective boundary” at this scale (responsible for possible diffusive ejection as discussed in §5) is increased only a little by the tight shearing.
Conclusion
==========
We have extended classical results from isoperimetric theory, concerned with identifying subsets of manifolds with least boundary size to volume ratios, to the situation where the manifolds are subjected to general nonlinear dynamics. We proved a dynamic version of (i) the Federer-Fleming Theorem, which tightly links geometric and functional characterisations of the fundamental isoperimetric problem, and (ii) the Cheeger inequality, which bounds the least boundary size to volume ratio by the first nontrivial eigenvalue of the Laplace operator on the manifold. We developed a new dynamic Laplace operator and used this operator to numerically identify subsets of manifolds that have small boundary size to volume ratios before, after, and during, the application of nonlinear dynamics. In nonlinear fluid flow, such sets characterise finite-time coherent sets, as their boundaries do not elongate and filament, and there is little exchange between the interior and exterior of these sets in the presence of small diffusion. We proved that the dynamic Laplace operator can also be obtained as a zero-diffusion limit of the existing probabilistic approach to identifying finite-time coherent sets [@F13], thus creating a strong formal link between probabilistic descriptions and geometric descriptions of Lagrangian coherent structures. Numerical experiments were carried out using a simple combination of Ulam’s method and a finite-difference scheme.
Obvious extensions of the methodology include handling non-volume-preserving dynamics, nonuniform initial mass distributions, and manifolds of nonvanishing curvature, and work is in progress in these directions. Accurate and efficient numerical methods are also being pursued. An advantage of the present formulation over [@F13] in the pure advection setting is that there is more freedom in selecting an approximating function basis as the basis no longer needs to generate numerical diffusion, and various out-of-the-box numerical methods can be employed. Recent work [@FJ15] uses radial basis functions to estimate both $\mathcal{P}$ and $\triangle$ and has resulted in a more accurate approximation of the eigenspectrum and a significant reduction of the number of required Lagrangian trajectories, compared to the numerical techniques in the present paper. Radial basis functions are flexible enough to be able to handle irregularly-shaped domains as sometimes arise in applications.
Acknowledgements
================
The author acknowledges feedback from Eric Kwok and Daniel Karrasch, which improved the manuscript, assistance from Oliver Junge regarding GAIO, and a discussion with Renato Feres. This research is supported by an Australian Research Council Future Fellowship and Discovery Project DP150100017.
Proof of Theorem \[fflemma\]
============================
\[sl\] Let $A\in GL(d)$, and $v_1,\ldots, v_d$ be an orthonormal basis for $\mathbb{R}^d$. Let $U_1=\sp\{v_1,\ldots, v_k\}, U_2=\sp\{v_{k+1},\ldots,v_{d}\}$. Then $$\|A(v_1\wedge \cdots \wedge v_k)\|=|\det(A)|\cdot\|(A^{-1})^\top(v_{k+1}\wedge\cdots \wedge v_d)\|,$$ where $\|\cdot\|$ is the volume induced by the Gram determinant.
The parallelopiped $A(v_1\wedge\cdots\wedge v_d)$ has volume $|\det(A)|$ by orthonormality of $v_1,\ldots,v_d$. We note that the space spanned by $(A^{-1})^\top v_{k+1},\ldots, (A^{-1})^\top v_d$ is orthogonal to the space spanned by $Av_1,\ldots A v_k$; indeed any element of one collection is orthogonal to any element of the other. The volume of the parallelopiped can therefore be written as $\det(A)=\|Av_1\wedge\cdots\wedge A v_k\|\cdot\|\Pr_{(A^{-1})^\top(U_2)}(Av_{k+1})\wedge\cdots\wedge \Pr_{(A^{-1})^\top(U_2)}(A v_d)\|$, where $\Pr_{(A^{-1})^\top(U_2)}$ denotes orthogonal projection along $A(U_1)$ onto $(A^{-1})^\top(U_2)$. Let $V$ be the $d\times (d-k)$ matrix with columns $v_{k+1},\ldots,v_d$, and let $W=(A^{-1})^\top V$. The projection matrix associated with $\Pr_{(A^{-1})^\top(U_2)}$ is $C=W(W^\top W)^{-1}W^\top$. We compute $\|\Pr_{(A^{-1})^\top(U_2)}(Av_{k+1})\wedge\cdots\wedge \Pr_{(A^{-1})^\top(U_2)}(A v_d)\|$ as $\det((CAV)^\top CAV)^{1/2}$. $$\begin{aligned}
\lefteqn{\det((CAV)^\top CAV)}\\
&=&\det\left(\left[V^\top A^\top (A^{-1})^\top V(V^\top A^{-1}(A^{-1})^\top V)^{-1}V^\top A^{-1}\right] \left[(A^{-1})^\top V(V^\top A^{-1}(A^{-1})^\top V)^{-1}V^\top A^{-1}AV\right]\right)\\
&=&\det\left(\left(V^\top A^{-1}(A^{-1})^\top V\right)^{-1}\left(V^\top A^{-1}(A^{-1})^\top V\right)\left(V^\top A^{-1}(A^{-1})^\top V\right)^{-1}\right)\quad\mbox{by orthogonality of $V$}\\
&=&\det(V^\top A^{-1}(A^{-1})^\top V)^{-1})\\
&=&1/\|(A^{-1})^\top (v_{k+1}\wedge\cdots\wedge v_d)\|^2\end{aligned}$$ Thus, $\|\Pr_{(A^{-1})^\top(U_2)}(Av_{k+1})\wedge\cdots\wedge \Pr_{(A^{-1})^\top(U_2)}(A v_d)\|=1/\|(A^{-1})^\top(v_{k+1}\wedge\cdots \wedge v_d)\|$, and the result follows.
The main thing to prove is the equality. We modify the arguments of Remark VI.2.3 and the proof of Theorem II.2.1[@chavelisoperimetric]. (a) We start by showing $\mathbf{s}^D\le \mathbf{h}^D$. We do this by creating a specific sequence of functions $f_\epsilon$, which when substituted into (\[soboleveqn\]), in the limit achieve $\mathbf{h}^D$; therefore $\mathbf{s}^D$ can potentially be lower still. Suppose we have a specific disconnection $\Gamma$, and define $\Gamma_\epsilon=\{x\in M: d(x,\Gamma)<\epsilon\}$, where $d(x,\Gamma)=\inf_{y\in \Gamma}\|x-y\|$, and because of the vanishing curvature we write the Riemannian distance between two points $x,y\in M$ as $\|x-y\|$. Define $$f_\epsilon=\left\{
\begin{array}{ll}
1, & \hbox{$x\in M_1\setminus\Gamma_\epsilon$;} \\
-1, & \hbox{$x\in M_2\setminus\Gamma_\epsilon;$}\\
(1/\epsilon)d(x,\Gamma),&\hbox{$x\in M_1\cap\Gamma_\epsilon$;}\\ -(1/\epsilon)d(x,\Gamma),&\hbox{$x\in M_2\cap\Gamma_\epsilon.$}
\end{array}
\right.$$ The function $f_\epsilon$ is Lipschitz and by mollification on $\mathring{M}$ we can produce a sequence of $C^\infty$ functions $\phi_{j,\epsilon}$ such that $\| f_\epsilon- \phi_{j,\epsilon}\|_1\to 0$ and $\|\nabla f_\epsilon-\nabla \phi_{j,\epsilon}\|_1\to 0$ as $j\to \infty$ (see e.g. Theorem I.3.3 [@chavelisoperimetric]). Now, $$\begin{aligned}
\mathbf{s}^D&=&\inf_{f\in C^\infty} \frac{\|\nabla f\|_1+\|\nabla\P f\|_1}{2\inf_\alpha \|f-\alpha\|_{1}}\\
&\le&\frac{\|\nabla \phi_{j,\epsilon}\|_1+\|\nabla\P \phi_{j,\epsilon}\|_1}{2\inf_\alpha \|\phi_{j,\epsilon}-\alpha\|_{1}}\qquad\mbox{for each $j$}\\
&\le&\frac{\|\nabla f_\epsilon\|_1+\|\nabla \phi_{j,\epsilon}-\nabla f_\epsilon\|_1+\|\nabla\P f_\epsilon\|_1+\|\nabla\P \phi_{j,\epsilon}-\nabla\P f_\epsilon\|_1}{2\inf_\alpha \|f_\epsilon-\alpha\|_{1}-2\|f_\epsilon-\phi_{j,\epsilon}\|_{1}}\qquad\mbox{for each $j$}\\\end{aligned}$$ Thus, letting $j\to\infty$ we have for each $\epsilon>0$, $$\label{sDcompare1}
\mathbf{s}^D\le \frac{\|\nabla f_\epsilon\|_1+\|\nabla\P f_\epsilon\|_1}{2\inf_\alpha \|f_\epsilon-\alpha\|_{1}}.$$ We begin to interpret these terms in terms of $d$- and $d-1$-dimensional volume. Note that $|\nabla f_\epsilon|$ is $1/\epsilon$ on $\Gamma_\epsilon$ and zero elsewhere. Thus $\lim_{\epsilon\to 0}\int_M |\nabla f_\epsilon|\ d\ell=\lim_{\epsilon\to 0}\ell(\Gamma_\epsilon)/\epsilon=2\ell_{d-1}(\Gamma)$.
Now we concentrate on the term $\|\nabla \P f_\epsilon\|_1$. Let $x\in\Gamma_\epsilon\cap M_2$, and $z\in\Gamma$ be the closest point to $x$ (if there are several, choose one). Note $\nabla f_\epsilon(x)=\hat{n}(x)/\epsilon$ where $\hat{n}(x)=(z-x)/|z-x|$, which is normal to $\Gamma$ at $z$. Since $T$ is volume-preserving we note that $\P f_\epsilon$ is 1 on $T( M_1\setminus\Gamma_\epsilon)$ and $-1$ on $T( M_2\setminus\Gamma_\epsilon)$. Thus, $|\nabla (\P f_\epsilon)|=0$ on these regions. The value of $\P f_\epsilon$ on $T\Gamma_\epsilon$ must be computed. Let us first consider $T( M_2\cap\Gamma_\epsilon)$. $$\begin{aligned}
\nonumber\int_{T( M_2\cap\Gamma_\epsilon)}|\nabla (f_\epsilon\circ T^{-1})(x)|\ d\ell&=&\int_{T( M_2\cap\Gamma_\epsilon)}|\nabla f_\epsilon(T^{-1}x)^\top\cdot DT^{-1}(x)|\ d\ell\\
\nonumber&=&(1/\epsilon)\int_{T( M_2\cap\Gamma_\epsilon)}|\hat{n}(T^{-1}x)^\top\cdot DT^{-1}(x)|\ d\ell\\
\nonumber&=&(1/\epsilon)\int_{ M_2\cap\Gamma_\epsilon}|\hat{n}(x)^\top\cdot DT^{-1}(Tx)|\ d\ell\\
\label{final}&=&(1/\epsilon)\int_{ M_2\cap\Gamma_\epsilon}|(DT(x)^{-1})^\top\hat{n}(x)|\ d\ell\end{aligned}$$
Let $t_1(x),\ldots,t_{d-1}(x)$ be an orthonormal set of vectors spanning the orthogonal complement of $\hat{n}(x)$ in $\mathbb{R}^d$ (these vectors span the $d-1$-dimensional tangent space of $\Gamma$ at $z$). By Lemma \[sl\], one has $|(DT(x)^{-1})^\top\hat{n}(x)|=|DT(x)(t_1(x)\wedge\cdots \wedge t_{d-1}(x))|$, where $|\cdot|$ denotes the volume (one-dimensional and $d-1$-dimensional, respectively) induced by the Gram determinant. Thus, $$(\ref{final})=(1/\epsilon)\int_{ M_2\cap\Gamma_\epsilon}|DT(x)(t_1(x)\wedge\cdots \wedge t_{d-1}(x))|\ d\ell.$$ The integrand measures the local increase in the $d-1$-dimensional volume of linear spaces close to the tangent spaces of $\Gamma$, under the action of $T$ in an $\epsilon$-neighbourhood of $\Gamma$, and the above integral converges to $\ell_{d-1}(T\Gamma)$ as $\epsilon\to 0$. Similarly, $$\lim_{\epsilon\to 0}\int_{T( M_1\cap\Gamma_\epsilon)}|\nabla (f_\epsilon\circ T^{-1})(x)|\ d\ell=\ell_{d-1}(T\Gamma).$$ Thus, $$\label{sDcompare2}
\lim_{\epsilon\to 0} (\|\nabla f_\epsilon\|_1+\|\nabla \P f_\epsilon\|_1)/2=\ell_{d-1}(\Gamma)+\ell_{d-1}(T\Gamma).$$
Now we turn to the denominator $\int_M |f_\epsilon-\alpha|\ d\ell$. Without loss, suppose that $\ell( M_1)\le \ell( M_2)$. $$\begin{aligned}
\int_M |f_\epsilon-\alpha|\ d\ell&\ge& |1-\alpha|(\ell( M_1)-\ell(\Gamma_\epsilon))+|1+\alpha|(\ell( M_2)-\ell(\Gamma_\epsilon))\\
&\ge&(|1-\alpha|+|1+\alpha|)(\ell( M_1)-\ell(\Gamma_\epsilon))\\
&\ge&2(\ell( M_1)-\ell(\Gamma_\epsilon)),\end{aligned}$$ implying $\inf_\alpha \int_M |f_\epsilon-\alpha|\ d\ell\ge 2(\ell( M_1)-\ell(\Gamma_\epsilon))$ for each $\epsilon>0$. Taking the limit as $\epsilon\to 0$, we combine this with (\[sDcompare1\]) and (\[sDcompare2\]) to conclude $\mathbf{s}^D\le \mathbf{h}^D$.
\(b) Now let $f\in C^\infty(M)$ and choose a constant $\beta$ so that, $ M_1=\{f>\beta\}$, $ M_2=\{f<\beta\}$ have equal volume. Such a choice of $\beta$ satisfies $\|f-\beta\|_1=\inf_\alpha \|f-\alpha\|_1$ (see Remark VI.2.2 p163 [@chavelisoperimetric]). For $t>0$ define $D_t=\{x\in M_1: f(x)>\beta+t\}$, and $\tilde{D}_t=\{x\in T(M_1): \P f(x)>\beta+t\}=
\{x\in T(M_1): f\circ T^{-1}(x)>\beta+t\}$, thus $\tilde{D}_t=TD_t$. In what follows, we concentrate on $\tilde{D}_t$ and $\P f$, modifying the argument for $D_t$ and $f$ in [@chavelisoperimetric] p46. Firstly, using the co-area formula, Corollary I.3.1 [@chavelisoperimetric] with $f\equiv 1$, $\Phi=f-\beta$ (and then $\Phi=\mathcal{P}f-\beta$), one has $$\label{coarea1}
\int_{ M_1}|\nabla (f -\beta)|\ d\ell+\int_{T(M_1)}|\nabla (\P f -\beta)|\ d\ell=\int_0^\infty (\ell_{d-1}(\partial{D}_t)+\ell_{d-1}(\partial\tilde{D}_t))\ dt.$$ Continuing, $$\begin{aligned}
\label{altsobvol}
(\ref{coarea1})&\ge& 2\mathbf{h}^D \int_0^\infty \ell(D_t)\ dt\quad\mbox{since $\ell(D_t)=\ell(\tilde{D}_t)\le \ell(M)/2$}\\
\label{hfinal}&=& 2\mathbf{h}^D \int_{ M_1} |f-\beta|\ d\ell,\end{aligned}$$ by a standard argument, see e.g. p.164 [@chavelisoperimetric]. Similarly, $\int_{ M_2}|\nabla (f -\beta)|\ d\ell+\int_{T(M_2)}|\nabla (\P f -\beta)|\ d\ell\ge 2\mathbf{h}^D \int_{ M_2} |f-\beta|\ d\ell$. Thus, $$\begin{aligned}
\label{altsob1}\int_{ M}|\nabla f|\ d\ell+\int_{T(M)}|\nabla \P f |\ d\ell&=&\int_{ M}|\nabla (f -\beta)|\ d\ell+\int_{T(M)}|\nabla (\P f -\beta)|\ d\ell\\
\nonumber&\ge& 2\mathbf{h}^D \int_{ M} |f-\beta|\ d\ell\\
\nonumber&\ge& 2\mathbf{h}^D \inf_\alpha \int_{ M} |f-\alpha|\ d\ell\end{aligned}$$ and $\mathbf{s}^D\ge \mathbf{h}^D$.
\(c) In order to get the inequality $\mathbf{h}^D/2\le \hat{\mathbf{s}}^D$, we set $\beta=\bar{f}$ in the argument of part (b) above. Note that now possibly only one of $M_1$ or $M_2$ has volume less than or equal to $\ell(M)/2$, and WLOG suppose it is $M_1$. Then (\[hfinal\]) holds. We note that $\int_{T(M_2)}|\mathcal{P}f-\beta|\ d\ell=\int_{M_2} |f-\beta|\ d\ell\ge \int_{M_1} |f-\beta|\ d\ell=\int_{T(M_1)}|\mathcal{P}f-\beta|\ d\ell$. Thus, $$\begin{aligned}
\frac{\int_M |\nabla(f-\beta)|\ d\ell +\int_{T(M)}|\nabla(\mathcal{P}f-\beta)|\ d\ell}{2\int_M|f-\beta|\ d\ell}&\ge&\frac{\int_{M_1} |\nabla(f-\beta)|\ d\ell +\int_{T(M_1)}|\nabla(\mathcal{P}f-\beta)|\ d\ell}{4\int_{M_1}|f-\beta|\ d\ell}\\
&\ge&\mathbf{h}^D/2,\end{aligned}$$ by (\[hfinal\]). Thus, $\mathbf{h}^D\le 2\hat{\mathbf{s}}^D$.
All of the calculations concerning the map $T$ and the operator $\mathcal{P}$ in the proof of Theorem \[ffthm\] hold for each of the maps $T^{(i)}, i=1,\ldots,n$ or $T^{(t)}, t\in[0,\tau]$. These calculations are always put together linearly, and the proof proceeds exactly as in the proof of Theorem \[ffthm\].
Proof of Theorem \[cheegerdynthm\]
==================================
The proof is a modification of the presentation of [@ledoux]; see also [@chavelisoperimetric] Theorem 3, Section IV.3. Let $g:M\to \mathbb{R}$ be positive and smooth; then $\P g$ is also positive and smooth. First, by the co-area formula applied separately to $g$ and $\P g$ (see e.g. Cor. I.3.1 [@chaveleigenvalues]) and then the definition of $\mathbf{h}^D$ we have that $$\begin{aligned}
\label{coareaeqn}\int_M |\nabla g|\ d\ell+\int_{T(M)}|\nabla(\P g)|\ d\ell&=&\int^\infty_0 \ell_{d-1}(\{g=t\})+\ell_{d-1}(\{\P g=t\})\ dt \\
&=&\int^\infty_0 \ell_{d-1}(\{g=t\})+\ell_{d-1}(T\{g=t\})\ dt \\
\label{heqn}&\ge& 2\mathbf{h}^D\int_0^\infty \min\{\ell(\{g\ge t\}),\ell(\{g<t\})\}\ dt\end{aligned}$$
Let $f:M\to \mathbb{R}$ be smooth and denote by $m$ the median of $f$; i.e. $\ell(f\ge m)\ge 1/2$ and $\ell(f\le m)\ge 1/2$. Set $f^+=\max\{f-m,0\}, f^-=\max\{-f+m,0\}$, so that $f-m=f^+-f^-$. Note that by volume-preservation, $m$ is also the median for $\P f$, and we similarly decompose $\P f-m=(\P f)^+-(\P f)^-$. Further, note that since $\P$ is positive and a composition operator, we have $((\P f)^+)^2=\P ((f^+)^2)$ and similarly for $f^-$. We apply (\[heqn\]) to $g=(f^+)^2$ and $g=(f^-)^2$. Note that for each $t>0$, $\ell(\{(f^+)^2\ge t\})\le 1/2$ and $\ell(\{(f^-)^2\ge t\})\le 1/2$. Now, $$\begin{aligned}
\nonumber\lefteqn{\frac{1}{2}\left(\int_M |\nabla((f-m)^2)|\ d\ell+\int_{T(M)}|\nabla((\P f-m)^2)|\ d\ell\right)}\\
\nonumber&=&\frac{1}{2}\left(\int_M |\nabla((f^+)^2)|+|\nabla((f^-)^2)|\ d\ell+\int_{T(M)}|\nabla((\P f^+)^2)|+|\nabla((\P f^-)^2)|\ d\ell\right)\\
\nonumber&\ge&\mathbf{h}^{D}\int^\infty_0 \ell(\{(f^+)^2\ge t\})\ dt+\mathbf{h}^{D}\int^\infty_0 \ell(\{(f^-)^2\ge t\})\ dt\\
\nonumber&=&\mathbf{h}^{D}\int_M (f^+)^2\ d\ell+\mathbf{h}^{D}\int_M (f^-)^2\ d\ell\\
\label{eqn1}&=&\mathbf{h}^{D}\int_M (f-m)^2\ d\ell\end{aligned}$$ Further, $$\begin{aligned}
\nonumber\frac{1}{2}\int_M|\nabla((f-m)^2)|\ d\ell&=&\int_M|(f-m)\cdot\nabla f|\ d\ell\\
\label{cauchyschwartz}&\le& \|f-m\|_2\cdot\|\nabla f\|_2,\end{aligned}$$ where $\|\cdot\|_2$ denotes the $L^2(\ell)$ norm. Also analogously to (\[cauchyschwartz\]) we have $$\label{cauchyschwartza}
\frac{1}{2}\int_{T(M)}|\nabla((\P f-m)^2)|\ d\ell\le \|\P f-m\|_2\cdot\|\nabla (\P f)\|_2=\|f-m\|_2\cdot\|\nabla (\P f)\|_2.$$ Thus, using (\[eqn1\])–(\[cauchyschwartza\]) and Cauchy-Schwartz, $$\begin{aligned}
\nonumber
(\mathbf{h}^{D})^2\|(f-m)\|_2^4&\le&\|f-m\|_2^2\left(\|\nabla f\|_2+\|\nabla(\P f)\|_2\right)^2\\
\label{penultimate}&\le&2\|f-m\|_2^2\left(\|\nabla f\|_2^2+\|\nabla(\P f)\|_2^2\right).\end{aligned}$$ Thus (\[penultimate\]) becomes $$\begin{aligned}
\nonumber(\mathbf{h}^{D})^2&\le &2\frac{\int_M|\nabla f|^2\ d\ell+\int_{T(M)}|\nabla(\P f)|^2\ d\ell}{\int_M (f-m)^2\ d\ell}\\
\label{cheegerexpr}&\le &
2\frac{\int_M|\nabla f|^2\ d\ell+\int_{T(M)}|\nabla(\P f)|^2\ d\ell}{\int_M (f-(\int_M f\ d\ell))^2\ d\ell},\end{aligned}$$ since $\inf_\alpha\|f-\alpha\|_2$ is realised when $\alpha$ is the mean of $f$. As $f$ is arbitrary, we may minimise the RHS of (\[penultimate\]) by inserting $f=u_2$, the eigenfunction of $\hat{\triangle}$ corresponding to the lowest nontrivial eigenvalue (with $u_2$ satisfying the boundary condition of Theorem \[cheegerdynthm\] if $M$ has nonempty boundary). Upon this insertion, the RHS of (\[penultimate\]) takes the value $4\lambda_2$ by Part 3 of Theorem \[dynlapthm\].
All of the calculations concerning the map $T$ and the operator $\mathcal{P}$ in the proof of Theorem \[cheegerdynthm\] hold for each of the maps $T^{(i)}, i=1,\ldots,n$ or $T^{(t)}, t\in[0,\tau]$. These calculations are almost always put together linearly; the only exception is equation (\[penultimate\]), which we describe here in the continuous time case. $$\begin{aligned}
\nonumber
(\mathbf{h}^{D})^2\|(f-m)\|_2^4&\le&\frac{1}{\tau^2}\left(\int_0^\tau 2\|f-m\|_2\cdot\|\nabla \P^{(t)}f\|_2\ dt\right)^2\\
\label{penultimatet}&\le&\frac{4\|f-m\|_2^2}{\tau^2}\int_0^\tau \|\nabla \P^{(t)}f\|_2^2\ dt\cdot\int_0^\tau 1\ dt.
$$ Thus $$(\mathbf{h}^{D})^2\le 4\frac{\frac{1}{\tau}\int_0^\tau \|\nabla \P^{(t)}f\|_2^2\ dt}{\|f-m\|^2_2}.$$ The proof proceeds exactly as in the proof of Theorem \[cheegerdynthm\], using Remark \[spectrummultistep\].
Proof of Theorem \[dynlapthm\]
==============================
Existence of weak solutions and variational characterisation of eigenvalues {#sec:weakexist}
---------------------------------------------------------------------------
Let $X=W^{1,2}(M)$, the Sobolev space of functions $u:M\to\mathbb{R}$ with square-integrable weak derivative. The space $X$ is a Hilbert space with the inner product $\langle u,v\rangle_X=\int_M \nabla u\cdot \nabla v+uv\ d\ell$. We will establish the existence of a set of weak solutions $u\in X$ to $$\label{weakeqn}
(1/2)\left(\int_M \nabla v\cdot \nabla u\ d\ell+\int_{T(M)} \nabla (\P v)\cdot \nabla (\P u)\ d\ell\right)=-\lambda \int_M v\cdot u\ d\ell\qquad\mbox{for all $v\in W^{1,2}$}.$$ We require certain extremisation properties and therefore for $u\in X$, we define the functionals $F(u)=F_1(u)+F_2(u)$, where $F_1(u)=(1/2)\int_M |\nabla u|^2\ d\ell$, $F_2(u)=(1/2)\int_{T(M)}|\nabla(\P u)|^2\ d\ell$ and $G(u)=\int_M u^2\ d\ell-1$. We look for $u$ which minimizes $F(u)$ subject to $G(u)=0$ ($\|u\|_2=1$). In the following, we consider only $F_2(u)$ as the corresponding results for $F_1(u)$ follow immediately by setting $T$ to the identity map and $\P$ the identity operator.
\[functionallemma\]
1. The functional $F_2:X\to\mathbb{R}$ is well-defined,
2. The derivative $F_2'(u)$ is linear and bounded (hence $F_2'(u)\in X^*$),
3. $F_2$ is differentiable, and
4. $u\mapsto F'_2(u)$ is continuous as a map from $X$ to $X^*$.
- $2F_2(u)=\int_{T(M)} \|\nabla(\P u)\|^2\ d\ell=\int_{T(M)} \|\nabla(u\circ T^{-1})\|^2\ d\ell=\int_{T(M)} \|(DT^{-1})^\top\cdot(\nabla u)\circ T^{-1}\|^2\ d\ell$. By compactness of $M$ and the fact $T$ is $C^\infty$ diffeomorphism onto $T(M)$, one may find a $C<\infty$ such that the previous expression is bounded above by $C\int\|\nabla u\|^2\ d\ell\le C\|u\|_X^2<\infty$. Thus, $F_2:X\to \mathbb{R}$ is well-defined.
- $$\begin{aligned}
F'_2(u)v&=&\lim_{h\to 0}\frac{F_2(u+hv)-F_2(u)}{h}\\
&=&\lim_{h\to 0}\frac{\int_{T(M)}|\nabla(\P u)+h\nabla (\P v)|^2\ d\ell-\int_{T(M)}|\nabla (\P u)|^2\ d\ell}{2h}\\
&=&\lim_{h\to 0}\frac{\int_{T(M)}|\nabla(\P u)|^2+2h(\nabla(\P u)\cdot\nabla (\P v))+h^2|\nabla (\P v)|^2 -|\nabla (\P u)|^2\ d\ell}{2h}\\
&=&\lim_{h\to 0}\frac{\int_{T(M)} 2h(\nabla(\P u)\cdot\nabla (\P v))+h^2|\nabla (\P v)|^2\ d\ell}{2h}\\
&=&\int_{T(M)} \nabla(\P u)\cdot\nabla (\P v)\ d\ell.\end{aligned}$$
$F'_2(u)$ is clearly linear in $v$ and bounded because $\int_{T(M)} \nabla(\P u)\cdot\nabla (\P v)\ d\ell\le \|\nabla(\P u)\|_2\cdot\|\nabla (\P v)\|_2\le C^{1/2}\|u\|_X\|v\|_X$, where $C$ is the constant from part (i); thus $F_2'(u)\in X^*$.
- $F_2$ is differentiable since $$\begin{aligned}
|F_2(u+v)-F_2(u)-F_2'(u)v|&=&\left|(1/2)\int_{T(M)}|\nabla(\P v)|^2\ d\ell\right|\\
&\le& (C/2)\int_M |\nabla v|^2\ d\ell\\
&\le& (C/2)\|v\|_X^2\to 0\mbox{ as }\|v\|_X\to 0.\end{aligned}$$
- Finally, let $u,v,w\in X$, then $$\begin{aligned}
|(F'_2(u)-F'_2(v))w|&=&\left|\int_{T(M)} (\nabla(\P u)-\nabla(\P v))\cdot\nabla(\P w)\ d\ell\right|\\
&\le& \|\nabla(\P(u-v))\|_2\cdot\|\nabla(\P w)\|_2\\
&\le& C\|\nabla(u-v)\|_2\cdot\|\nabla w\|_2\\
&\le& C\|u-v\|_X\cdot\|w\|_X.\end{aligned}$$ Thus $\|F'_2(u)-F'_2(v)\|_{X^*}=\sup_{\|w\|_X=1}|(F'_2(u)-F'_2(v))w|\to 0$ as $\|u-v\|_X\to 0$, and $F'_2:X\to X^*$ is continuous.
\[lemmaattain\] $F$ attains its minimum on the constraint set $\mathcal{C}=\{u\in X: G(u)=0\}$.
Let $I=\inf_{u\in X}\{F(u):G(u)=0\}$. Select a sequence $u_j\in\mathcal{C}$ such that $F(u_j)\to I$ and $F(u_j)\le I+1$ for all $j\ge 0$. By the Poincaré inequality (e.g. p163 [@mcowen]), the norm $\|\cdot\|_X$ is equivalent to $\|\nabla(\cdot)\|_2$, so using the form of $F$, the $u_j$ are uniformly bounded in norm in $X$. By standard arguments using Rellich compactness (e.g. Thm. 8.4.2 [@jost]), one can find a subsequence $u_{j_k}\in X$ and a function $\bar{u}\in X$ such that $u_{j_k}\to \bar{u}$ in $L^2(M)$. For the weak derivatives, we develop a Cauchy sequence. We first demonstrate that there is a $c>0$ such that $\|\nabla ( u_{j_k}-u_{j_l})\|\le (1/c)\|\nabla(\mathcal{P}u_{j_k}-\mathcal{P}u_{j_l})\|$. $$\begin{aligned}
\|\nabla(\mathcal{P}u_{j_k}-\mathcal{P}u_{j_l})\|^2&=&\int_{T(M)}|\nabla(u_{j_k}\circ T^{-1})-\nabla(u_{j_k}\circ T^{-1})|^2\ d\ell\\
&=&\int_{T(M)}|\nabla(u_{j_k}-u_{j_k})\circ T^{-1}\cdot DT^{-1}(x)|^2\ d\ell\\
&=&\int_{M}|\nabla(u_{j_k}-u_{j_k})\cdot DT^{-1}(Tx)|^2\ d\ell\\
&\ge& c^2\int_M |\nabla(u_{j_k}-u_{j_k})|^2\ d\ell,\end{aligned}$$ where $c=\inf_{x\in M, v\in\mathbb{R}^d} |DT^{-1}(Tx)\cdot v|/|v|>0$ as $T$ is a $C^\infty$ diffeomorphism and $M$ is compact. Now, $$\begin{aligned}
\nonumber\lefteqn{(1+c^2)\|\nabla u_{j_k}-\nabla u_{j_l}\|^2}\\
\nonumber&\le& \|\nabla u_{j_k}-\nabla u_{j_l}\|^2+\|\nabla (\P u_{j_k})-\nabla (\P u_{j_l})\|^2\\
\nonumber&=&2\left(\|\nabla u_{j_k}\|^2+\|\nabla (\P u_{j_k})\|^2\right)+2\left(\|\nabla u_{j_l}\|^2+\|\nabla (\P u_{j_l})\|^2\right)-\left(\|\nabla(u_{j_k}+u_{j_l})\|^2+\|\nabla(\P(u_{j_k}+u_{j_l}))\|^2\right)\\
\label{expansion}
&\le&2\left(\|\nabla u_{j_k}\|^2+\|\nabla (\P u_{j_k})\|^2\right)+2\left(\|\nabla u_{j_l}\|^2+\|\nabla (\P u_{j_l})\|^2\right)-2I\|(u_{j_k}+u_{j_l})\|^2\end{aligned}$$ By construction, as $j_k,j_l\to\infty$, the first two terms of (\[expansion\]) both converge to $4I$, and the final term of (\[expansion\]) converges to $8I$. Thus, the $u_{j_k}$ form a Cauchy sequence in $W^{1,2}$, and converge to $\bar{u}$ in $W^{1,2}$.
Because $F$ and $G$ are both[^4] $C^1$ functionals on $X$, we can use the method of Lagrange multipliers, and by Lemma \[lemmaattain\] the minimiser $\bar{u}$ satisfies the Euler-Lagrange equation $F'(\bar{u})v=\mu G'(\bar{u})v$ for some $\mu\in\mathbb{R}$ and all $v\in X$. By the constructions in the proof of Lemma \[functionallemma\] (ii), this equation is $$\label{weakeigeneqn}
\int_M (\nabla \bar{u}\cdot\nabla v)\ d\ell+\int_{T(M)}(\nabla(\P \bar{u})\cdot \nabla(\P v))\ d\ell=-2\mu \int_M \bar{u}v\ d\ell\quad\mbox{for all $v\in X$}.$$ If we set $\lambda=\mu$, we have exactly the statement (\[weakeqn\]). In fact, putting $v=\bar{u}$ we get $I=F(\bar{u})=(1/2)\int_M |\nabla \bar{u}|^2 +|\nabla(\P \bar{u})|^2\ d\ell=-\lambda\int_M \bar{u}^2\ d\ell=-\lambda$. Thus we could have defined $\lambda$ by the Rayleigh quotient $$-\lambda=\inf_{u\in X}\left(\frac{\int_M |\nabla u|^2\ d\ell +\int_{T(M)}|\nabla(\P u)|^2\ d\ell}{2\int_M u^2\ d\ell}\right).$$ From now on we denote $(\lambda,\bar{u})$ by $(\lambda_1,u_1)$ and search for other solution pairs. Note that $-\lambda_1=F(u_1)\ge 0$, but that $u_1\equiv 1$ yields $F(u_1)=0$ by volume-preservation of $T$; thus $\lambda_1=0$.
\[orthogonality\] If $(\lambda_1,u_1)$ and $(\lambda_2,u_2)$ are solution pairs for (\[weakeqn\]) with $\lambda_1\neq\lambda_2$ then $\langle u_1,u_2\rangle =0$; that is $u_1,u_2$ are orthogonal in the $L^2$ inner product.
Put $u=u_1, v=u_2$ in (\[weakeqn\]), then put $u=u_2, v=u_1$ in (\[weakeqn\]) and subtract the two equations to get $(\lambda_1-\lambda_2)\int_M u_1u_2\ d\ell=0$
One may now follow the standard procedure for constructing a sequence of eigenvalues $0=\lambda_1> \lambda_2>\cdots$ (e.g. [@mcowen] pp212–213) by first defining $$\label{rayleigh2}
-\lambda_2=\inf_{u\in X,\langle u,u_1\rangle=0}\frac{\int_M |\nabla u|^2\ d\ell+\int_{T(M)}|\nabla(\P u)|^2\ d\ell}{2\int_M u^2\ d\ell},$$ and then inductively adding the constraint $\langle u,u_2\rangle=0$ in the next infimum to define $\lambda_3$ and so on. The functions $u_1,u_2,\ldots,$ constructed in this way are scaled to form an orthonormal set in $L^2$.
\[infinitelemma\] The sequence $\lambda_n$ tends to $-\infty$ and for each $n$, the dimension of the solution space is finite.
Let $u_n,u_m$ be solutions to (\[rayleigh2\]) corresponding to $\lambda_n,\lambda_m$ obtained inductively as above. By (\[weakeqn\]), setting $u=u_n, v=u_m$, we have $$\label{infinite}
(1/2)\int_M \nabla u_n\cdot\nabla u_m\ d\ell+\int_{T(M)}\nabla(\P u_n)\cdot\nabla(\P u_m)\ d\ell=-\lambda_n\int u_nu_m\ d\ell.$$ By Lemma \[orthogonality\] we see that the RHS of (\[infinite\]) is 0 if $n\neq m$ and $\lambda_n$ if $n=m$. Thus, $$\|u_n\|_X^2=\int_M|\nabla u_n|^2 + u_n^2\ d\ell\le 2\left((1/2)\int_M|\nabla u_n|^2 \ d\ell+ \int_{T(M)}|\nabla (\P u_n)|^2\ d\ell\right)+1=-2\lambda_n+1.$$ By a standard argument, (e.g. [@mcowen] p213), we assume that $\lambda_n\nrightarrow-\infty$, thus $\|u_n\|_X$ is uniformly bounded in $n$ and by Rellich compactness one finds a Cauchy subsequence in $L^2$ and derives a contradiction by $L^2$ pairwise orthogonality of members of this subsequence.
Because $\lambda_n\to-\infty$, each $\lambda_n$ occurs only finitely many times and therefore the solution space for each $\lambda_n$ is finite-dimensional.
Ellipticity and strong solutions {#sec:ellipticity}
--------------------------------
We have established the existence of a set of solutions $(u_1,\lambda_1), (u_2,\lambda_2),\ldots,$ of (\[weakeqn\]) and now wish to make a link between the solutions of (\[weakeqn\]) and solutions of the strong formulation (\[strongeqn\]). The property of ellipticity of $\hat{\triangle}$ will be crucial. Suppose we have a second order differential operator $$\label{2ndorderop}
L=\sum_{i,j=1}^d a_{ij}(x)\frac{\partial^2}{\partial x_i\partial x_j}+\sum_{i=1}^d b_i(x)\frac{\partial}{\partial x_i}+c(x),\quad x\in \mathring{M},$$ with coefficient functions $a_{ij}, b_i, c$ that are $C^\infty$ on $M$. We will say that $a_{ij}$ satisfies uniform ellipticity if $$\label{ellipticity}
\sum_{i,j=1}^d a_{ij}(x)\xi_i\xi_j\ge \gamma|\xi|^2,\qquad\mbox{for all $x\in M$, $\xi\in \mathbb{R}^d$}.$$
\[ellipticlemma\] $\hat{\triangle}$ satisfies uniform ellipticity.
We have $L=\hat{\triangle}=\triangle+\P^*\triangle\P$. Clearly $\triangle$ is elliptic, with $a_{ij}(x)=\delta_{ij}$ for all $x\in M$. We show that $\P^*\triangle\P$ is elliptic and the result follows. In fact, since $\P^*$ is merely composition with $T$, we only need show that $\triangle\P$ is elliptic.
Let $x_1,\ldots,x_d$ denote the standard coordinate system on $M$ in which $\triangle f=\sum_{i=1}^d \frac{\partial^2 f}{\partial x_i^2}$. Let $f:M\to\mathbb{R}$ be $C^2$; denoting $T^{-1}(x_1,\ldots,x_d)=(T^{-1}_1(x_1,\ldots,x_d),\ldots,T^{-1}_d(x_1,\ldots,x_d))$, and applying the chain rule for partial differentiation, one has $$\begin{aligned}
\lefteqn{
\frac{\partial^2}{\partial x_ix_j}\left(f\circ T^{-1}\right)}\\
&=&\left[\frac{\partial T^{-1}_1}{\partial x_j},\ldots,\frac{\partial T^{-1}_d}{\partial x_j}\right]\cdot H(f)\circ T^{-1}\cdot\left[\frac{\partial T^{-1}_1}{\partial x_i},\ldots,\frac{\partial T^{-1}_d}{\partial x_i}\right]^\top+(\nabla f)\circ T^{-1}\cdot\left[\frac{\partial^2 T^{-1}_1}{\partial x_ix_j},\ldots,\frac{\partial^2 T^{-1}_d}{\partial x_ix_j}\right]^\top,\end{aligned}$$ where $H(f)$ is the Hessian for $f$. As $\triangle(\P f)=\sum_{i=1}^d \frac{\partial^2}{\partial x_i^2}(f\circ T^{-1})$, the representation of $\triangle (\P f)$ in the form (\[2ndorderop\]) is $$\label{2ndorderopspecific}
\sum_{i,k,l=1}^d \frac{\partial T^{-1}_k}{\partial x_i}\frac{\partial T^{-1}_l}{\partial x_i}[H(f)]_{kl}\circ T^{-1}+\sum_{i,k=1}^d \frac{\partial^2 T^{-1}_k}{\partial x_i^2}[\nabla f]_k\circ T^{-1}.$$ Since $T$ is a $C^\infty$ diffeomorphism, both $a_{kl}=\sum_{j=1}^d\frac{\partial T^{-1}_k}{\partial x_j}\frac{\partial T^{-1}_l}{\partial x_j}$ and $b_i=\sum_{j=1}^d\frac{\partial^2 T^{-1}_i}{\partial x_j^2}$ are $C^\infty$ and bounded as a function of $x\in T(M)$ for each $k,l=1,\ldots,d$. In order to show that $a_{kl}$ is uniformly elliptic, we note that $a_{kl}(x)$ is the inner product of the $k^{th}$ and $l^{th}$ rows of the Jacobian matrix $D(T^{-1})(x)$. Thus, for each $x\in T(M)$, $a_{kl}(x)$ is a Gram matrix, formed from the $d$ (linearly independent, because $T$ is a diffeomorphism) rows of $D(T^{-1})(x)$, denoted $r_1,\ldots,r_d$. The matrix $a_{kl}(x)$ is symmetric and therefore positive definite if and only if all of its eigenvalues are positive (e.g. Theorem 7.2.1 [@hornjohnson]). Using the structure of the Gram matrix, we know $a_{kl}(x)$ is positive semidefinite and is nonsingular if and only if $\{r_1,\ldots,r_d\}$ are linearly independent (e.g. Theorem 7.2.10 [@hornjohnson]). Thus, $a_{kl}(x)$ is positive definite and satisfies $\sum_{k,l=1}^d a_{kl}(x)\xi_k\xi_l\ge \gamma(x)|\xi|^2$ for some $\gamma(x)>0$ for every $x\in M$. By compactness of $M$, we can set $\gamma=\min_{x\in M}\gamma(x)$.
We show that solutions of (\[weakeqn\]) are in $C^\infty(M)$ and satisfy (\[strongeqn\])–(\[strongbc\]). By the arguments used to obtain Corollary 8.4.1 [@jost] or the discussion on p.214 [@GilbargTrudinger], provided that our second-order differential operator $\hat{\triangle}$ is uniformly elliptic and that $T$ and $M$ are $C^\infty$, one has that a solution $u$ of (\[weakeqn\]) is in fact $C^\infty$ on $M$. Following the arguments in §8.4–8.5 [@jost], because $u\in C^\infty(M)$ we can apply Green’s first identity to the first term on the LHS of (\[weakeqn\]) to obtain: $$\label{green1}
\int_M \nabla v\cdot \nabla u\ d\ell=-\int_M v (\triangle u)\ d\ell+\int_{\partial M} v (\nabla u)\cdot \mathbf{n}\ d\ell_{d-1}.$$ Now, the second term on the LHS of (\[weakeqn\]): denoting $\tilde{\mathbf{n}}(y)$ to be the outward unit normal at $y\in T(M)$, and using Green’s first identity and change of variables under $T$: $$\begin{aligned}
\int_{T(M)} \nabla (\P v)\cdot \nabla (\P u)\ d\ell&=&-\int_{T(M)} \P v\cdot \triangle \P u\ d\ell+\int_{\partial T(M)} \P v \left[\nabla (\P u)\right]\cdot \tilde{\mathbf{n}}\ d\ell_{d-1}\\
&=&-\int_M v\cdot(\P^*\triangle\P)u\ d\ell+\int_{\partial T(M)} \P v \left[\nabla (\P u)\right]\cdot \tilde{\mathbf{n}}\ d\ell_{d-1}.\end{aligned}$$ We now manipulate the second term above using the chain rule for differentiation, change of variables under $T$, and volume-preservation of $T$: $$\begin{aligned}
\nonumber
\int_{\partial T(M)} \P v \left[\nabla (\P u)\right]\cdot \tilde{\mathbf{n}}\ d\ell_{d-1}&=&\int_{\partial T(M)} v\circ T^{-1} \left[\left((\nabla u)\circ T^{-1}\right)^\top\cdot DT^{-1}\right]\cdot \tilde{\mathbf{n}}\ d\ell_{d-1}\\
\label{halfwayeqn}&=&\int_{\partial M} v \left[\left(\nabla u\right)^\top \cdot (DT^{-1}\circ T)\right]\cdot (\tilde{\mathbf{n}}\circ T)|\det DT_{|\mathcal{T}_x(\partial M)}|\ d\ell_{d-1}.
$$ Note that $\tilde{\mathbf{n}}(Tx)=(DT(x)^{-1})^\top \mathbf{n}(x)/\|(DT(x)^{-1})^\top \mathbf{n}(x)\|$. By Lemma \[sl\], $|\det DT_{|\mathcal{T}_x(\partial M)}|=\|(DT(x)^{-1})^\top \mathbf{n}(x)\|$. Thus, $$(\ref{halfwayeqn})=\int_{\partial M} v \left[\left(\nabla u\right)^\top \cdot(DT)^{-1} \right]\cdot \left((DT)^{-1}\right)^\top\cdot\mathbf{n}\ d\ell_{d-1},$$ and we arrive at the transformed version of (\[weakeqn\]): $$\begin{aligned}
\label{weakeqnsmooth1}
\lefteqn{(1/2)\int_M v\cdot(\triangle+\P^*\triangle\P)u\ d\ell=\lambda \int_M v\cdot u\ d\ell}\\
\label{weakeqnsmooth2}&&\quad+(1/2)\int_{\partial M} v (\nabla u)\cdot \mathbf{n}\ d\ell_{d-1}+(1/2)\int_{\partial M} v \left[\left(\nabla u\right)^\top \cdot(DT)^{-1} \right]\cdot \left((DT)^{-1}\right)^\top\cdot\mathbf{n}\ d\ell_{d-1}, \forall v\in W^{1,2}.\end{aligned}$$ By considering all $v\in W^{1,2}_0$ (the closure of $C^\infty_0(\mathring{M})\cap W^{1,2}(M)$ with respect to $\|\cdot\|_{W^{1,2}(M)}$) in (\[weakeqnsmooth1\])–(\[weakeqnsmooth2\]) we see that $(1/2)(\triangle+\P^*\triangle\P)u=\lambda u$ on $\mathring{M}$ (let $f=(1/2)(\triangle+\P^*\triangle\P)u-\lambda u$ and WLOG suppose $f(x)>0$ at some $x\in\mathring{M}$. Necessarily, $f(x)>0$ for $x$ in some open $O\subset \mathring{M}$ and consider a bump function $v$ positive in a ball contained in $O$ and zero outside $O$ to derive a contradiction). Now (\[weakeqnsmooth2\]) implies that $$\int_{\partial M} v (\nabla u)\cdot \mathbf{n}\ d\ell_{d-1}+\int_{\partial M} v \left[\nabla u \cdot (DT)^{-1}\cdot \left((DT)^{-1}\right)^\top\right]\cdot\mathbf{n}\ d\ell_{d-1}=0\quad \forall v\in W^{1,2}.$$ Again using the fact that $u\in C^\infty(M)$, by an argument on $\partial M$ analogous to the parenthetical argument above, it follows that (\[strongbc\]) holds.
Proof of Theorem \[objthm\]
---------------------------
A key component to the proof of Theorem \[objthm\] is the fact that the Laplacian commutes with isometries.
\[isometrylemma\] $$\label{isometryeqn}
\triangle(f\circ \Phi_{t_0})=(\triangle f)\circ\Phi_{t_0}.$$
By (\[2ndorderopspecific\]), and the fact that $\Phi_{t_0}$ is affine, the LHS of (\[isometryeqn\]) is $$\begin{aligned}
\sum_{i,k,l=1}^d Q(t_0)_{ki}\cdot[H(f)]_{kl}\circ \Phi_{t_0} \cdot Q(t_0)_{li}
&=&\operatorname{Tr}(Q(t_0)^\top\cdot [H(f)]\circ \Phi_{t_0}\cdot Q(t_0))\\
&=&\operatorname{Tr}([H(f)]\circ \Phi_{t_0}),\end{aligned}$$ using orthogonality of $Q(t_0)$, to obtain exactly the RHS of (\[isometryeqn\]).
We first show equivalence of (\[origevalprob\]) and (\[transfevalprob\]). $$\begin{aligned}
\dot{\hat{\triangle}}&=&\triangle+\P^*_{\dot{T}}\triangle \P_{\dot{T}}\\
&=&\triangle+(\P_{\Phi_{t_1}}\circ \P\circ \P_{\Phi_{t_0}}^{-1})^*\triangle(\P_{\Phi_{t_1}}\circ \P\circ \P_{\Phi_{t_0}}^{-1})\\
&=&\triangle+\P_{\Phi_{t_0}}\P^*\triangle\P \P_{\Phi_{t_0}}^{-1},\end{aligned}$$ where we have used Lemma \[isometrylemma\] and the fact that $\P_{\Phi_{t_0}}$ and $\P_{\Phi_{t_1}}$ are unitary operators. Now, if $\hat{\triangle}f=\lambda f$, $$\begin{aligned}
\dot{\hat{\triangle}}\P_{\Phi_{t_0}}f&=&(\triangle \P_{\Phi_{t_0}}+\P_{\Phi_{t_0}}\P^*\triangle\P)f\\
&=&\P_{\Phi_{t_0}}(\triangle +\P^*\triangle\P)f\\
&=&\lambda\P_{\Phi_{t_0}}f,\end{aligned}$$ as required, where we have again used Lemma \[isometrylemma\].
We now demonstrate equivalence of (\[strongorigbc\]) and (\[strongtransfbc\]). We calculate each of the components of (\[strongtransfbc\]). First, note that $\nabla (\P_{\Phi_{t_0}}f)(x)=\left[(\nabla f)\circ\Phi_{t_0}^{-1}(x)\right]^\top\cdot Q(t_0)^{-1}$. Next, $D\dot{T}(x)=Q(t_1)\cdot DT(\Phi^{-1}_{t_0}(x))\cdot Q(t_0)^{-1}$, so $D\dot{T}(x)^{-1}=Q(t_0)\cdot DT(\Phi^{-1}_{t_0}(x))^{-1}\cdot Q(t_1)^{-1}$ and $\left(D\dot{T}(x)^{-1}\right)^\top=Q(t_1)\cdot \left(DT(\Phi^{-1}_{t_0}(x))^{-1}\right)^\top\cdot Q(t_0)^{-1}$. Thus, for $x\in \partial(\Phi_{t_0}(M))$, $$\begin{aligned}
\lefteqn{(\ref{strongtransfbc})}\\
&=&(\nabla f)\circ\Phi_{t_0}^{-1}(x)\cdot Q(t_0)^{-1}\cdot[Q(t_0)\mathbf{n}(\Phi_{t_0}^{-1}x)+Q(t_0)\cdot DT(\Phi_{t_0}^{-1}x)^{-1}\cdot \left(DT(\Phi_{t_0}^{-1}x)^{-1}\right)^\top \mathbf{n}(\Phi_{t_0}^{-1}x)]\\
&=&(\nabla f)\circ\Phi_{t_0}^{-1}(x)\cdot[\mathbf{n}(\Phi_{t_0}^{-1}x)+DT(\Phi_{t_0}^{-1}x)^{-1}\cdot \left(DT(\Phi_{t_0}^{-1}x)^{-1}\right)^\top \mathbf{n}(\Phi_{t_0}^{-1}x)],\end{aligned}$$ which is exactly condition (\[strongorigbc\]) evaluated at $\Phi_{t_0}^{-1}x\in \partial M$.
Proof of Theorem \[analcvgce\] {#sec:analcvgce}
==============================
The crux of the proof of Theorem \[analcvgce\] is linking the diffusion operators $\mathcal{D}_{M,\epsilon}, \mathcal{D}_{T(M),\epsilon}$ with $\triangle$. This linking is possible because of the symmetry of the smoothing kernel $q_\epsilon$. At small scales (small $\epsilon$), $\mathcal{D}_{M,\epsilon}, \mathcal{D}_{T(M),\epsilon}$ are close to the identity operator, and because of the spatial symmetry of $q_\epsilon$, the next dominant term depends on second order derivatives.
\[nddifflemma\] Let $M$ be a connected, compact Riemannian manifold of vanishing curvature, and $f:M \to\mathbb{R}$ be $C^3$. Let $q:M\to\mathbb{R}^+$ be a nonnegative density with compact support, with mean the origin, and covariance matrix $c\cdot I$, where $I$ is the $d\times d$ identity matrix. We scale $q$ to form $q_\epsilon(x)=q(x/\epsilon)/\epsilon^d$, and define $\D f(x)=\int_M q_\epsilon(x-y)f(y)\ d\ell(y)$. Then $$\label{diffform1}\lim_{\epsilon\to 0} \frac{(\D-I)f(x)}{\epsilon^2}=(c/2)\triangle f(x),$$ for each $x\in \mathring{M}$.
Using Taylor, we expand $f$ in an $\epsilon$-ball about $x$ $$\label{taylor}
f(x+y)=\sum_{|\alpha|=0}^2\frac{D^\alpha f(x)}{|\alpha| !} y^\alpha+\sum_{|\alpha|=3} R_\alpha(x+y)y^\alpha,$$ where $R_\alpha(x+y)$ is the remainder term. The notation used is $\alpha=(\alpha_1,\ldots,\alpha_d)$, where $\alpha_i$ is the number of derivatives in coordinate direction $x_i$; $|\alpha|$ denotes the sum of the elements of $\alpha$ and $\alpha !=\prod_{i=1}^d \alpha_i !$. If $\partial M\neq \emptyset$, then put $\epsilon'=\inf_{z\in\partial M}\dist(x,z)$. In the following we assume that $\epsilon\le \epsilon'$. $$\begin{aligned}
\nonumber\D f(x)&=&\int_M q_\epsilon(x-y)f(y)\ d\ell(y)\\
\nonumber&=&\int_M q_\epsilon(y)f(x+y)\ d\ell(y) \\
\label{Reqn}&=&\int_M q_\epsilon(y)\left[\sum_{|\alpha|=0}^2\frac{D^\alpha f(x)}{|\alpha| !} y^\alpha+\sum_{|\alpha|=3} R_\alpha(x+y)y^\alpha\right]\ d\ell(y)\end{aligned}$$ The terms in the first sum of order $|\alpha|=0, 1, 2$, respectively, are: $f(x)$, 0, $\sum_{|\alpha|=2} m_\alpha(q_\epsilon) D^\alpha f(x)/\alpha !$, where $m_\alpha(q_\epsilon)=\int_M q_\epsilon(y) y^\alpha\ d\ell(y)$ denotes the tensor of $\alpha$-moments of $q_\epsilon$. The order 1 term is zero because of the assumption on the mean (the $|\alpha|=1$ moments) of $q$. We note that $m_\alpha(q_\epsilon)=\int_M q_\epsilon(y) y^\alpha\ d\ell(y)=\int_M q(y/\epsilon)/\epsilon^d y^\alpha\ d\ell(y)=\int_M q(z)\epsilon^{|\alpha|}z^\alpha\ d\ell(z)=\epsilon^{|\alpha|}m_\alpha(q)$. We can further simplify the order 2 term as $\sum_{|\alpha|=2} m_\alpha(q_\epsilon) D^\alpha f(x)/\alpha !=(c/2)\epsilon^2\triangle f(x)$ using the fact that the covariance matrix of $q$ is $c\times I$. Rearranging (\[Reqn\]), we have $$\D f(x)-f(x)-(c/2)\epsilon^2\triangle f(x)
=\int_M q_\epsilon(y)\sum_{|\alpha|=3}R_\alpha(x+y)y^\alpha\ d\ell(y).$$ For $|\alpha|=3$, and $y\in B_\epsilon(x)$, one has $|R_\alpha(x+y)|\le (1/\alpha !)\max_{|\alpha|=3}\max_{z\in B_\epsilon(x)}|D^\alpha f(z)|=:C(x)$, so $$|\D f(x)-f(x)-(c/2)\epsilon^2\triangle f(x)|\le C(x)\epsilon^3 \sum_{|\alpha|=3} m_\alpha(q).$$
By Lemma \[nddifflemma\], we have $\D f(x)=f(x)+\epsilon^2(c/2)\triangle f(x)+O(\epsilon^3)$, where $O(\epsilon^3)$ means the error term is of the form $\epsilon^3 \mathcal{R}(x)$. Therefore $\P \D f(x)=\P f(x)+\epsilon^2(c/2)\P \triangle f(x)+O(\epsilon^3)$. Since $T$ is $C^3$ we may again apply Lemma \[nddifflemma\] to obtain $$\begin{aligned}
\D\P\D f(x)&=&\D\left(\P f(x)+\epsilon^2(c/2)\P \triangle f(x)+O(\epsilon^3)\right)\\
&=&\P f(x)+\epsilon^2((c/2)\P\triangle f(x)+(c/2)\triangle\P f(x))+O(\epsilon^3)\end{aligned}$$ Finally, $$\begin{aligned}
\D^*\P^*\D^*\D\P\D f(x)&=&\P^*\P f(x)+\epsilon^2(c/2)(\P^*\P\triangle+\P^*\triangle \P+\P^*\triangle \P+\triangle\P^*\P)f(x)+O(\epsilon^3)\\
&&=f(x)+\epsilon^2 c(\triangle+\P^*\triangle \P)f(x)+O(\epsilon^3)
$$
\[4thorderremark\] It is reasonably natural for $q$ to have additional symmetry, so that $m_\alpha(q)=0$ for $|\alpha|=3$. In this case, the error term in the above proof is $O(\epsilon^4)$. For example, the uniform diffusion on a unit ball considered in Example \[unifexample\] has $m_\alpha(q)=0$ for all $|\alpha|$ odd.
[10]{}
M.R. Allshouse and J.-L. Thiffeault. Detecting coherent structures using braids. , 241(2):95–105, 2012.
L. Arnold. . Springer, 1998.
S. Balasuriya and C.K.R.T. Jones. Diffusive draining and growth of eddies. , 8(4/5):241–251, 1999.
I. Chavel. , volume 115 of [*Pure and Applied Mathematics*]{}. Academic Press, Orlando, 1984.
I. Chavel. , volume 145 of [*Cambridge Tracts in Mathematics*]{}. Cambridge University Press, 2001.
J. Cheeger. A lower bound for the smallest eigenvalue of the [L]{}aplacian. , 625:195–199, 1970.
M. Dellnitz, G. Froyland, and O. Junge. . In B. Fiedler, editor, [*Ergodic Theory, Analysis, and Efficient Simulation of Dynamical Systems*]{}, pages 145–174. Springer, 2001.
H. Federer and W.H. Fleming. Normal and integral currents. , pages 458–520, 1960.
G. Froyland. An analytic framework for identifying finite-time coherent sets in time-dependent dynamical systems. , 250:1–19, 2013.
G. Froyland, C. Horenkamp, V. Rossi, N. Santitissadeekorn, and A. Sen Gupta. Three-dimensional characterization and tracking of an [A]{}gulhas [R]{}ing. , 52–53:69–75, 2012.
G. Froyland, C. Horenkamp, V. Rossi, and Erik van Sebille. Studying an Agulhas ring’s long-term pathway and decay with finite-time coherent sets. To appear in *Chaos*.
G. Froyland and O. Junge. On fast computation of finite-time coherent sets using radial basis functions. *Chaos*, 25:087409, 2015.
G. Froyland and K. Padberg-Gehle. Almost-invariant and finite-time coherent sets: directionality, duration, and diffusion. In Wael Bahsoun, Chris Bose, and Gary Froyland, editors, [*Ergodic Theory, Open Dynamics, and Coherent Structures*]{}, volume 70 of [ *Proceedings in Mathematics and Statistics*]{}, chapter 9, pages 171–216. Springer, 2014.
G. Froyland, N. Santitissadeekorn, and A. Monahan. Transport in time-dependent dynamical systems: Finite-time coherent sets. , 20:043116, 2010.
D. Gilbarg and N.S. Trudinger. . Classics in Mathematics. Springer, Berlin, revised third printing of second edition, 2001.
D.S. Grebenkov and B.-T. Nguyen. Geometrical structure of [L]{}aplacian eigenfunctions. , 55(4):601–667, 2014.
G. Haller. A variational theory of hyperbolic [L]{}agrangian coherent structures. , 240:574––598, 2011.
G. Haller and F.J. Beron-Vera. Geodesic theory of transport barriers in two-dimensional flows. , 241:1680––1702, 2012.
R.A. Horn and C.R. Johnson. . Cambridge University Press, 1985.
. . Number 214 in Graduate Texts in Mathematics. Springer, New York, 2002.
M. Ledoux. A simple analytic proof of an inequality by [P. Buser]{}. , 121(3):951–959, 1994.
T. Ma and E. Bollt. Differential geometry perspective of shape coherence and curvature evolution by finite-time nonhyperbolic splitting. , 13(3):1106–1136.
J.A.J. Madrid and A.M. Mancho. Distinguished trajectories in time dependent vector fields. , 19(1):013111, 2009.
V.G. Mazya. Classes of domains and imbedding theorems for function spaces. In [*Soviet Math. Dokl*]{}, volume 1, pages 882–885, 1960.
R.C. McOwen. . Prentice-Hall, Upper Saddle River, 1996.
I. Mezi[ć]{}, S. Loire, V.A. Fonoberov, and P. Hogan. A new mixing diagnostic and gulf oil spill movement. , 330(6003):486–489, 2010.
B.A. Mosovsky and J.D. Meiss. Transport in transitory dynamical systems. , 10(1):35–65, 2011.
R. Mundel, E. Fredj, H. Gildor, and V. Rom-Kedar. New [L]{}agrangian diagnostics for characterizing fluid flow mixing. , 26:126602, 2014.
J. M. Ottino. . Cambridge University Press, 1989.
R.T. Pierrehumbert and H. Yang. Global chaotic mixing on isentropic surfaces. , 50(15):2462–2480, 1993.
C. Truesdell and W. Noll. . Springer, 3rd edition, 2004.
S. Ulam. . Interscience, 1964.
[^1]: Frequently, *coherent structures* are co-dimension 1 objects, while full-dimensional objects are called *coherent sets*.
[^2]: To weaken the smoothness assumption on $F(x,\cdot)$, but still obtain smooth flow maps, see [@arnold] Appendix B.3.
[^3]: in [@F13] there is an additional normalisation term required for non-Lebesgue reference measures and non-Lebesgue-preserving $T$. In the present paper, as $T$ is volume preserving and volume is our reference measure, we eschew this normalisation term here.
[^4]: Showing $G$ is a $C^1$ functional follows identically to the arguments above for $F$.
|
{
"pile_set_name": "ArXiv"
}
|
INTRODUCTION {#sec1-1}
============
Cerebrospinal puncture is an important procedure for diagnosis of some critical conditions such as subarachnoid hemorrhage and central nervous system infections. Also, therapeutic lumbar puncture (LP) is indicated in benign intracranial hypertension.\[[@ref1][@ref2]\]
LP is routinely carried out with the patient in the supine or lateral decubitus position using two techniques. Traditional technique uses Tuffier\'s line (an imaginary line connecting both iliac crests) to determine the level of the puncture site. This line crosses L4-L5 space. LP is usually performed in this space or 1 level over this, on the L3-L4 space.\[[@ref3]\] Although this technique has high success rate it depends on personal expertise to find interspinous space. Inability to find the proper space is associated with repetitive painful attempts and high possibility of traumatized tap.\[[@ref4]\]
An alternate method to surface landmark guided LP is ultrasound (US) guidance. For many years, US has been used to facilitate this procedure.\[[@ref5]\] The use of US to find puncture site was first described 40 years ago in a medical journal in Russian language.\[[@ref6]\]
In this technique vertebral column and surrounding structures can be shown. By placing US probe vertically and horizontally on spinous process of vertebras, subdural space would be seen as a dark space inside the spinal column. To find the most appropriate point for needle insertion, the examiner must be able to determine five structures by US: Spinous process, ligamentum flavum, dura, epidural, and subarachnoid spaces.\[[@ref7][@ref8]\]
Ten years later some anesthesiologists used US for better visualization of epidural and subarachnoid spaces for instillation of anesthetic drugs.\[[@ref9]\] Also, in a series of studies, researchers showed that US could increase successful attempts and easiness of the procedure especially in obese patients.\[[@ref10]\]
Interestingly, primary training of US-guided LP could improve expertise of physicians to identify pertinent landmarks to facilitate the procedure.\[[@ref11][@ref12]\] Although US-guided LP is not a new technique, the same studies in the field of emergency medicine are limited, especially in some residents with short training program of sonography.
The primary objective of this study was to compare the efficacy and feasibility of US-guided LP with traditional method in patients admitted to ED. We also assessed the performance of US-guided LP regarding to body mass index (BMI) of patients.
MATERIALS AND METHODS {#sec1-2}
=====================
Study design {#sec2-1}
------------
This study was a prospective randomized clinical trial that was registered in Iranian Registry of Clinical Trials by code number of IRCT201008172337N4. It was carried out in the academic Emergency Department (ED) of Hazrat Rasoul Akram hospital, Tehran, Iran. The study was approved by the Local Ethics Committee.
Patients {#sec2-2}
--------
We enrolled patients who admitted to our ED from March 2009 to March 2010 and were candidate for LP because of any medical problems \[[Diagram 1](#F1){ref-type="fig"}\]. Exclusion criteria were age less than 18 or more than 60, mass in the brain, local infection at the puncture level, pregnant women, coagulopathy, previous spinal surgery, and those with unwillingness to participate in the study.
{#F1}
Intervention {#sec2-3}
------------
Patients were included to undergo LP using US guided or palpation of surface landmarks by simple randomization method. In the US group, insertion point determined at the L3-L4 inter-vertebral space by use of a 7.5-10 MHz linear probe. For deeper penetration in used. Sonographic landmarks were spinous processes, dura mater, ligamentum flavum, and epidural space \[[Figure 1](#F2){ref-type="fig"}\].
{#F2}
In the control group, puncture level was determined by palpation of external landmarks, anterior superior iliac spines, and the spinous process of lumbar vertebrae L3, L4, and L5 which was performed by the study physicians. LPs were performed in a sterile manner with the patient in lateral decubitus fetal position.
Three PG-3 (post-graduate year) emergency medicine residents who underwent 2 h training on US to learn relevant spinal landmarks performed the procedures by supervision of a board-certified emergency medicine attending physician.
Before starting the study, all three residents performed some practice scans in healthy volunteers to ensure that they could capture the best possible image.
Measurements {#sec2-4}
------------
Demographic characteristics, pain score using numerical rating scale from 0 to 10, number of attempts (needle insertion), procedure time (time to obtain the best scan possible until emersion of cerebrospinal fluid), and number of traumatic LP (needle-induced blood in the cerebrospinal fluid) were recorded by another observing emergency medicine resident. BMI was defined as the individual\'s body mass (in kilograms) divided by the square of his or her height (in meters). According to BMI, patients were categorized into three groups: Normal, less than 25; overweight, 25-29; obese, more than 29.
Sample size and statistic analysis {#sec2-5}
----------------------------------
Sample size was estimated in 80 cases (40 cases in each group) by applying Altman\'s Nomogram. The procedure time was the variable that sample size was calculated based on it. The sample size was nearly similar to previous studies.\[[@ref13][@ref14][@ref15]\] It was calculated considering the power of 80%, significance level of 0.05, and the standardized difference of 0.6. Informed consent was obtained from patients who were included in this study.
A statistical analysis was performed by means of the Mann--Whitney U-test, Student\'s *t*-test. Chi-squared test used when nominal data were compared between two groups. Data from BMI groups were compared using two-way analysis of variance and Bonferroni *post hoc* test. Two-way analysis of variance (ANOVA) for BMI and time of procedure, number of attempts, number of traumatic LP, and pain score were performed for each procedure subgroups. When the ANOVA was significant, a Bonferroni test was then used for multiple comparisons. All tests were two-tailed, and a *P* value of 0.05 was accepted as the limit of significance. Data analysis was performed using the Statistical Package for the Social Sciences (SPSS, Version 14, IBM, Chicago, Illinois, USA).
RESULTS {#sec1-3}
=======
In this clinical trial, 80 eligible patients was allocated to two groups, each consisted of 40 cases. Among them, 48 patients (60%) were females. The patients' mean of age was 42.3 ± 3.5 years old (44.09 ± 3.4 in US group and 40.51 ± 4.1 in landmark group). Patients categorized into three groups regarding BMI: 12 patients (15%) with BMI of less than 25, 42 patients (52.5%) with BMI of 25-29, and 26 patients (32.5%) with BMI more than 29. There was no statistically significant difference between two groups from the point of sex, mean of age, and BMI (*P* \> 0.05). Details are showed in [Table 1](#T1){ref-type="table"}.
######
Characteristics of participants based on methods of lumbar puncture
In all of the patients LP was performed successfully. Patients who underwent US-guided LP reported lower pain score (4.4 ± 1.4 vs. 7.4 ± 1.1, *P* = 0.001). Numbers of attempts and numbers of traumatic LP were less in US-guided group too (*P* = 0.047 and *P* = 0.024 respectively). US-guided LPs were performed faster and the mean of procedure times in the US group were 3.3 ± 1.2 min. In control group this time was 6.4 ± 1.2 min. (*P* = 0.032). Results are demonstrated in [Table 2](#T2){ref-type="table"}.
######
Comparison of different aspects of lumbar puncture regarding on ultrasound or traditional methods
A two-way ANOVA model with interaction was used to evaluate the association of time, numbers of attempts, numbers of traumatic LP, and pain score with BMI subgroups and procedure group. Deviations differed between BMI subgroups \[[Table 3](#T3){ref-type="table"}\]. As shown, the durations of doing LP are markedly longer in patients with different BMI subgroups in traditional method (ANOVA, Bonferroni test, *P* = 0.000). Again participants with different BMI subgroups underwent less traumatic LP when US-guided method used (two-sided ANOVA, Bonferroni test, *P* = 0.035). Number of attempts and patient\'s pain scoring differed significantly in patients with different BMI subgroups and was significantly higher in traditional group (two-sided ANOVA, Bonferroni test, *P* = 0.001). Actually, US-guided LP improved all the factors assessed in our participants with different BMI subgroups and especially, BMI more than 29.
######
Two-way ANOVA model associations between procedure and body mass index subgroups, with time of procedure, number of attempts, procedure time, and pain score
DISCUSSION {#sec1-4}
==========
LP is one of the important procedures in the EDs to obtain cerebrospinal fluid. The two different methods to find the insertion site are blind surface landmark guidance and US-guided techniques. Surface landmark guided LP has been shown to have high success rate in determining puncture site, but it may be difficult in some patients such as those with high BMI. Recently, US imaging of the spinal column has been proposed to facilitate identification of the pertinent landmarks for appropriate intervertebral space.\[[@ref16][@ref17]\]
Anesthesiologists have affirmed the use of US to find appropriate point of insertion for anesthetic catheters.\[[@ref18][@ref19]\] Coley *et al*.\[[@ref20]\] showed that US can diagnose the cause of unsuccessful LP. Moreover, they concluded that US can help determine whether or not to try further. In this study all of the procedures in both groups were successfully carried out (zero failure rates). Indeed, there is no relation between selections of each method. But we found a marked reduction in time need for performing procedure in US group. Besides, number of attempts and pain score in US group was significantly lower than control group especially, in patients with BMI more than 29.
On 2004, Peterson *et al*.\[[@ref21]\] concluded that use of US to guide LP had great benefit as a time-saving tool in ED. He showed that attending emergency physicians could easily perform US-guided LP when blind technique of needle insertion using surface landmark guidance was difficult or unsuccessful.
In the present study the number of traumatic LP in US group was significantly lower than control group (5 vs. 18). On 2007, a team of emergency physicians designed a study to compare the success rate and ease of performing LP in obese patients, with US-guided assistance or palpation of surface landmarks. They reported that using US could facilitate the procedure but there were no statistical differences in the number of attempts, patient comfort, traumatic LPs, or procedure time.\[[@ref10]\] Because that was the only study that compared the rate of traumatic taps between two techniques, it seemed that further prospective randomized controlled studies should be planned to determine this probable benefit of US-guided LP.
In the current study, we found that patients with BMI more than 29 had significantly better outcome in terms of pain score, procedure time, number of attempts, and traumatic taps. Stiffler\[[@ref15]\] compared the pain scores in patients with different BMI who underwent LP. He concluded that US-guided LP may be helpful in identifying pertinent landmarks in obese patients with BMI more than 30. These results have been reaffirmed in other studies.\[[@ref11][@ref14]\]
The current study evaluated the outcome of performing LP by PG-3 emergency medicine residents with two different methods. We recommend that future studies compare these methods in residents of other specialties such as neurology, internal medicine, pediatrics, and anesthesiology.
Notwithstanding the obvious benefits of US in determining proper insertion site in anesthetic blocks, there are few studies about the same profits for LP in EDs, so further studies in this area are needed.
CONCLUSION {#sec1-5}
==========
The current study showed that US-guided LP has better results in comparison with surface landmark guided method. These benefits are more apparent in patients who had BMI more than 29.
**Source of Support:** Nil
**Conflict of Interest:** None declared.
|
{
"pile_set_name": "PubMed Central"
}
|
Japan will not send warships to join a US-led maritime force to guard oil tankers in the Strait of Hormuz fearing a military response from Iran, but it may send patrol aircraft, said the Mainichi newspaper, citing unidentified government sources, Reuters reports.
But Japan may send warships independently to protect Japanese ships in the world’s most important oil artery, the newspaper said on Friday.
“We are closely monitoring the situation and continue to collect information while working closely with the United States and other countries,” said Japan’s Chief Cabinet Secretary Yoshihide Suga, when asked about the report.
As its key Asian ally and a major regional naval power, Washington is keen for Japan, which is the world’s fourth-biggest oil buyer, to play a major role in its proposed maritime force.
Press: South Korea to send naval unit to Strait of Hormuz
Japan’s government would likely face opposition at home to any military venture that could put its Self Defence Forces in harms way or threaten the well being of Japanese living in Iran. Japan’s military has not fought overseas since World War Two.
The United States has blamed Iran for a series of attacks since mid-May on shipping around the Strait of Hormuz, including one on a tanker operated by a Japanese shipping company. Tehran rejects the allegations.
US Secretary of State Mike Pompeo last week called on Japan, Britain, France, Germany, South Korea, Australia and other nations to join a maritime force to guard oil tankers sailing through the Strait of Hormuz.
|
{
"pile_set_name": "OpenWebText2"
}
|
Share text messages, inboxes, contacts, templates, lists, & notes with your team. Works with your computer & smartphone
CRM for text messaging from your web browser! Share messages, contacts, notes, lists, and templates with your team.
Get a dedicated number for your team to receive and send text messages with customers. Text messaging for business.
*** Featured on TechCrunch http://tcrn.ch/28ZXAQC ***
Heymarket provides an easy way for you and your team to manage the text messages shared with customers. Best of all, customers won't need to install anything!
SHARED INBOX
* Get a shared inbox so anyone on your team can instantly reply to customers.
* Assign conversations to specific team members.
* Create a custom signature for each team member.
SEND SMARTER MESSAGES
* Organize contacts into lists and send them private, personalized text messages. Each text is treated as a 1-on-1 message - like a Bcc: email.
* Schedule text messages to send them later.
* Templates allow you to reuse messages. Include text, photos, links, and even the first name of the contact receiving the message.
* Work in sync with your computer using Heymarket Web or your Android phone or iPhone.
* Share contacts, lists, templates with your team.
STAY PRODUCTIVE WHILE MESSAGING
* Never forget to follow up with a contact. Set a reminder for any conversation.
* Simple scheduling. With just a few taps, share your available dates and times in a text message.
* Archive and export text messages.
* Do not disturb to control when you receive work-related notifications.
* Add and share notes about each contact.
* Share contacts over text messaging.
* Annotate photos and share them.
NOTHING FOR YOUR CUSTOMERS TO INSTALL
* Get a professional identity with a second phone number.
* Unlimited picture and text messaging.
* Customers receive your messages as normal SMS/MMS on their phone.
*Use Heymarket for SMS Customer Outreach, SMS Support, SMS for Sales, SMS for Operations.
LEARN MORE
* Website - http://heymarket.com
* Blog - http://blog.heymarket.com
* Twitter - twitter.com/heymarketapp
* Facebook - facebook.com/heymarketapp
* Terms of Service - http://heymarket.com/tos
* Privacy Policy - http://heymarket.com/privacy
Read more
Read less
Rate this app
Write a review
0 / 2,000
Submit review
Delete
Edit
Reviews
English (United States)
All languages
Recent
Helpful
A Google User
August 9, 2017
Reply
This app is ideal if you are communicating with your customers over text messaging. We got a business text messaging number and our entire team can send and receive messages from it. We add notes to conversations we have with customers and the ability to assign conversations make it easy to get everyone involved. We also use it to send out text blasts to customers to keep them posted on upcoming events. The CRM features are ideal for what we need. Highly recommended!
|
{
"pile_set_name": "Pile-CC"
}
|
<!DOCTYPE html>
<!--[if lt IE 7 ]><html class="ie ie6" lang="en"> <![endif]-->
<!--[if IE 7 ]><html class="ie ie7" lang="en"> <![endif]-->
<!--[if IE 8 ]><html class="ie ie8" lang="en"> <![endif]-->
<!--[if (gte IE 9)|!(IE)]><!-->
<html lang="en" xmlns="http://www.w3.org/1999/html"> <!--<![endif]-->
<head>
<!-- Basic Page Needs
================================================== -->
<meta charset="utf-8" />
<title>icon-building: Font Awesome Icons</title>
<meta name="description" content="Font Awesome, the iconic font designed for Bootstrap">
<meta name="author" content="Dave Gandy">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<!--<meta name="viewport" content="initial-scale=1; maximum-scale=1">-->
<!--[if lt IE 9]>
<script src="http://html5shim.googlecode.com/svn/trunk/html5.js"></script>
<![endif]-->
<!-- CSS
================================================== -->
<link rel="stylesheet" href="../../assets/css/site.css">
<link rel="stylesheet" href="../../assets/css/pygments.css">
<link rel="stylesheet" href="../../assets/font-awesome/css/font-awesome.css">
<!--[if IE 7]>
<link rel="stylesheet" href="../../assets/font-awesome/css/font-awesome-ie7.css">
<![endif]-->
<!-- Le fav and touch icons -->
<link rel="shortcut icon" href="../../assets/ico/favicon.ico">
<script type="text/javascript" src="//use.typekit.net/wnc7ioh.js"></script>
<script type="text/javascript">try{Typekit.load();}catch(e){}</script>
<script type="text/javascript">
var _gaq = _gaq || [];
_gaq.push(['_setAccount', 'UA-30136587-1']);
_gaq.push(['_trackPageview']);
(function() {
var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true;
ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js';
var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s);
})();
</script>
</head>
<body data-spy="scroll" data-target=".navbar">
<div class="wrapper"> <!-- necessary for sticky footer. wrap all content except footer -->
<div class="navbar navbar-inverse navbar-static-top hidden-print">
<div class="navbar-inner">
<div class="container">
<a class="btn btn-navbar" data-toggle="collapse" data-target=".nav-collapse">
<span class="icon-bar"></span>
<span class="icon-bar"></span>
<span class="icon-bar"></span>
</a>
<a class="brand" href="../../"><i class="icon-flag"></i> Font Awesome</a>
<div class="nav-collapse collapse">
<ul class="nav">
<li class="hidden-tablet "><a href="../../">Home</a></li>
<li><a href="../../get-started/">Get Started</a></li>
<li class="dropdown-split-left"><a href="../../icons/">Icons</a></li>
<li class="dropdown dropdown-split-right hidden-phone">
<a href="#" class="dropdown-toggle" data-toggle="dropdown">
<i class="icon-caret-down"></i>
</a>
<ul class="dropdown-menu pull-right">
<li><a href="../../icons/"><i class="icon-flag icon-fixed-width"></i> Icons</a></li>
<li class="divider"></li>
<li><a href="../../icons/#new"><i class="icon-shield icon-fixed-width"></i> New Icons in 3.2.1</a></li>
<li><a href="../../icons/#web-application"><i class="icon-camera-retro icon-fixed-width"></i> Web Application Icons</a></li>
<li><a href="../../icons/#currency"><i class="icon-won icon-fixed-width"></i> Currency Icons</a></li>
<li><a href="../../icons/#text-editor"><i class="icon-file-text-alt icon-fixed-width"></i> Text Editor Icons</a></li>
<li><a href="../../icons/#directional"><i class="icon-hand-right icon-fixed-width"></i> Directional Icons</a></li>
<li><a href="../../icons/#video-player"><i class="icon-play-sign icon-fixed-width"></i> Video Player Icons</a></li>
<li><a href="../../icons/#brand"><i class="icon-github icon-fixed-width"></i> Brand Icons</a></li>
<li><a href="../../icons/#medical"><i class="icon-medkit icon-fixed-width"></i> Medical Icons</a></li>
</ul>
</li>
<li class="dropdown-split-left"><a href="../../examples/">Examples</a></li>
<li class="dropdown dropdown-split-right hidden-phone">
<a href="#" class="dropdown-toggle" data-toggle="dropdown">
<i class="icon-caret-down"></i>
</a>
<ul class="dropdown-menu pull-right">
<li><a href="../../examples/">Examples</a></li>
<li class="divider"></li>
<li><a href="../../examples/#new-styles">New Styles</a></li>
<li><a href="../../examples/#inline-icons">Inline Icons</a></li>
<li><a href="../../examples/#larger-icons">Larger Icons</a></li>
<li><a href="../../examples/#bordered-pulled">Bordered & Pulled</a></li>
<li><a href="../../examples/#buttons">Buttons</a></li>
<li><a href="../../examples/#button-groups">Button Groups</a></li>
<li><a href="../../examples/#button-dropdowns">Button Dropdowns</a></li>
<li><a href="../../examples/#bulleted-lists">Bulleted Lists</a></li>
<li><a href="../../examples/#navigation">Navigation</a></li>
<li><a href="../../examples/#form-inputs">Form Inputs</a></li>
<li><a href="../../examples/#animated-spinner">Animated Spinner</a></li>
<li><a href="../../examples/#rotated-flipped">Rotated & Flipped</a></li>
<li><a href="../../examples/#stacked">Stacked</a></li>
<li><a href="../../examples/#custom">Custom CSS</a></li>
</ul>
</li>
<li><a href="../../whats-new/">
<span class="hidden-tablet">What's </span>New</a>
</li>
<li><a href="../../community/">Community</a></li>
<li><a href="../../license/">License</a></li>
</ul>
<ul class="nav pull-right">
<li><a href="http://blog.fontawesome.io">Blog</a></li>
</ul>
</div>
</div>
</div>
</div>
<div class="jumbotron jumbotron-icon">
<div class="container">
<div class="info-icons">
<i class="icon-building icon-6"></i>
<span class="hidden-phone">
<i class="icon-building icon-5"></i>
<span class="hidden-tablet"><i class="icon-building icon-4"></i> </span>
<i class="icon-building icon-3"></i>
<i class="icon-building icon-2"></i>
</span>
<i class="icon-building icon-1"></i>
</div>
<h1 class="info-class">
icon-building
<small>
<i class="icon-building"></i> ·
Unicode: <span class="upper">f0f7</span> ·
Created: v3.0 ·
Categories:
Web Application Icons
</small>
</h1>
</div>
</div>
<div class="container">
<section>
<div class="row-fluid">
<div class="span9">
<p>After you get <a href="../../integration/">up and running</a>, you can place Font Awesome icons just about anywhere with the <code><i></code> tag:</p>
<div class="well well-transparent">
<div style="font-size: 24px; line-height: 1.5em;">
<i class="icon-building"></i> icon-building
</div>
</div>
<div class="highlight"><pre><code class="html"><span class="nt"><i</span> <span class="na">class=</span><span class="s">"icon-building"</span><span class="nt">></i></span> icon-building
</code></pre></div>
<br>
<div class="lead"><i class="icon-info-sign"></i> Looking for more? Check out the <a href="../../examples/">examples</a>.</div>
</div>
<div class="span3">
<div class="info-ad"><div id="carbonads-container"><div class="carbonad"><div id="azcarbon"></div><script type="text/javascript">var z = document.createElement("script"); z.type = "text/javascript"; z.async = true; z.src = "http://engine.carbonads.com/z/32291/azcarbon_2_1_0_VERT"; var s = document.getElementsByTagName("script")[0]; s.parentNode.insertBefore(z, s);</script></div></div>
</div>
</div>
</div>
</section>
</div>
<div class="push"><!-- necessary for sticky footer --></div>
</div>
<footer class="footer hidden-print">
<div class="container text-center">
<div>
<i class="icon-flag"></i> Font Awesome 3.2.1
<span class="hidden-phone">·</span><br class="visible-phone">
Created and Maintained by <a href="http://twitter.com/davegandy">Dave Gandy</a>
</div>
<div>
Font Awesome licensed under <a href="http://scripts.sil.org/OFL">SIL OFL 1.1</a>
<span class="hidden-phone">·</span><br class="visible-phone">
Code licensed under <a href="http://opensource.org/licenses/mit-license.html">MIT License</a>
<span class="hidden-phone hidden-tablet">·</span><br class="visible-phone visible-tablet">
Documentation licensed under <a href="http://creativecommons.org/licenses/by/3.0/">CC BY 3.0</a>
</div>
<div>
Thanks to <a href="http://maxcdn.com"><i class="icon-maxcdn"></i> MaxCDN</a> for providing the excellent <a href="http://www.bootstrapcdn.com/#tab_fontawesome">BootstrapCDN for Font Awesome</a>
</div>
<div class="project">
<a href="https://github.com/FortAwesome/Font-Awesome">GitHub Project</a> ·
<a href="https://github.com/FortAwesome/Font-Awesome/issues">Issues</a>
</div>
</div>
</footer>
<script src="http://platform.twitter.com/widgets.js"></script>
<script src="../../assets/js/jquery-1.7.1.min.js"></script>
<script src="../../assets/js/ZeroClipboard-1.1.7.min.js"></script>
<script src="../../assets/js/bootstrap-2.3.1.min.js"></script>
<script src="../../assets/js/site.js"></script>
</body>
</html>
|
{
"pile_set_name": "Github"
}
|
Q:
'find' all .log files under a directory, copy them to another directory named after their parent directory?
Random e.g.:
/stack-exchange/
|
|-- stack-overflow/
| |-- questions/
| |-- people/
| |-- site.log
|-- super-user/
| |-- questions/
| |-- people/
| |-- answers.log
|-- ask-ubuntu/
| |-- questions/
| |-- people/
| |-- linux.log
|-- logs/
This is the command I'd use to find all log files in a given directory:
find /stack-exchange/ -type f -name "*.log"
Now, going by the example, I'd like to copy all 3 log files (site.log, answers.log & linux.log) to /stack-exchange/logs directory and have them named after their parent directory, like so:
/stack-exchange/
|
|-- [...]
|
|-- logs/
| |-- stack-overflow.log
| |-- super-user.log
| |-- ask-ubuntu.log
How do I do that? I've tried this:
find /stack-exchange/ -type f -iname "*.log" -exec cp "{}" "../logs/$(basename "$(dirname "{}")").log" \;
I think I am close, but it doesn't work. All it does is create a ..log file. What am I doing wrong, and how do I fix it?
A:
Based on this answer to the question Use current filename ("{}") multiple times in "find -exec"?, this works:
find /stack-exchange/ -type f -iname "*.log" -exec sh -c 'cp "$0" "./logs/$(basename "$(dirname "$0")").log"' {} \;
Thank you @rdupz (see comment), for pointing me in the right direction.
|
{
"pile_set_name": "StackExchange"
}
|
No marketing agency staffer feels good when they see a retail client getting reviews like this on the web.
But we can find out why they’re happening, and if we’re going above-and-beyond in our work, we just might be able to catalyze turning things around if we’re committed to being honest with clients and have an actionable strategy for their in-store improvements.
In this post, I’ll highlight some advice from an internal letter at Tesla that I feel is highly applicable to the retail sector. I’d also like to help your agency combat the retail blues headlining the news these days with big brands downsizing, liquidating and closing up shop — I’m going to share a printable infographic with some statistics with you that are almost guaranteed to generate the client positivity so essential to making real change. And, for some further inspiration, I’d like to offer a couple of anecdotes involving an Igloo cooler, a monk, reindeer moss, and reviews.
The genuine pain of retail gone wrong: The elusive cooler, "Corporate," and the man who could hardly stand
“Hi there,” I greeted the staffer at the customer service counter of the big department store. “Where would I find a small cooler?”
“We don’t have any,” he mumbled.
“You don’t have any coolers? Like, an Igloo cooler to take on a picnic to keep things cold?”
“Maybe over there,” he waved his hand in unconcern.
And I stood there for a minute, expecting him to actually figure this out for me, maybe even guide me to the appropriate aisle, or ask a manager to assist my transaction, if necessary. But in his silence, I walked away.
“Hi there,” I tried with more specificity at the locally owned general store the next day. “Where would I find something like a small Igloo cooler to keep things cold on a picnic?”
“I don’t know,” the staffer replied.
“Oh…” I said, uncomfortably.
“It could be upstairs somewhere,” he hazarded, and left me to quest for the second floor, which appeared to be a possibly-non-code-compliant catch-all attic for random merchandise, where I applied to a second dimly illuminated employee who told me I should probably go downstairs and escalate my question to someone else.
And apparently escalation was necessary, for on the third try, a very tall man was able to lift his gaze to some coolers on a top shelf… within clear view of the checkout counter where the whole thing began.
Why do we all have experiences like this?
“Corporate tells us what to carry” is the almost defensive-sounding refrain I have now received from three employees at two different Whole Foods Markets when asking if they could special order items for me since the Amazon buyout.
Because, you know, before they were Amazon-Whole Foods, staffers would gladly offer to procure anything they didn’t have in stock. Now, if they stop carrying that Scandinavian vitamin D-3 made from the moss eaten by reindeer and I’ve got to have it because I don’t want the kind made by irradiating sheep wool, I’d have to special order an entire case of it to get my hands on a bottle. Because, you know, “Corporate.”
Why does the distance between corporate and customer make me feel like the store I’m standing in, and all of its employees, are powerless? Why am I, the customer, left feeling powerless?
So maybe my search for a cooler, my worries about access to reindeer moss, and the laughable customer service I’ve experienced don’t signal “genuine pain.” But this does:
This is genuine pain. When customer service is failing to the point that badly treated patrons are being further distressed by the sight of fellow shoppers meeting the same fate, the cause is likely built into company structure. And your marketing agency is looking at a bonafide reputation crisis that could presage things like lawsuits, impactful reputation damage, and even closure for your valuable clients.
When you encounter customer service disasters, it begs questions like:
Could no one in my situation access a list of current store inventory, or, barring that, seek out merchandise with me instead of risking the loss of a sale? Could no one offer to let “corporate” know that I’m dissatisfied with a “customer service policy” that would require me to spend $225 to buy a whole case of vitamins? Why am I being treated like a warehouse instead of a person? Could no one at the pharmacy see a man with a leg wound about to fall over, grab a folding chair for him, and keep him safe, instead of risking a lawsuit?
I think a “no” answer to all three questions proceeds from definite causes. And I think Tesla CEO, Elon Musk, had such causes in mind when he recently penned a letter to his own employees.
“It must be okay for people to talk directly and just make the right thing happen.”
“Communication should travel via the shortest path necessary to get the job done, not through the 'chain of command.' Any manager who attempts to enforce chain of command communication will soon find themselves working elsewhere.
A major source of issues is poor communication between depts. The way to solve this is allow free flow of information between all levels. If, in order to get something done between depts, an individual contributor has to talk to their manager, who talks to a director, who talks to a VP, who talks to another VP, who talks to a director, who talks to a manager, who talks to someone doing the actual work, then super dumb things will happen. It must be ok for people to talk directly and just make the right thing happen.
In general, always pick common sense as your guide. If following a 'company rule' is obviously ridiculous in a particular situation, such that it would make for a great Dilbert cartoon, then the rule should change.”
- Elon Musk, CEO, Tesla
Let’s parlay this uncommon advice into retail. If it’s everyone’s job to access a free flow of information, use common sense, make the right thing happen, and change rules that don’t make sense, then:
Inventory is known by all store staff, and my cooler can be promptly located by any employee, rather than workers appearing helpless. Employees have the power to push back and insist that, because customers still expect to be able to special order merchandise, a specific store location will maintain this service rather than disappoint consumers. Pharmacists can recognize that patrons are often quite ill and can immediately place some chairs near the pharmacy counter, rather than close their eyes to suffering.
“But wait,” retailers may say. “How can I trust that an employee’s idea of ‘common sense’ is reliable?”
Let’s ask a monk for the answer.
“He took the time...”
I recently had the pleasure of listening to a talk given by a monk who was defining what it meant to be a good leader. He hearkened back to his young days, and to the man who was then the leader of his community.
“He was a busy man, but he took the time to get to know each of us one-on-one, and to be sure that we knew him. He set an example for me, and I watched him,” the monk explained.
Most monasteries function within a set of established rules, many of which are centuries old. You can think of these guidelines as a sort of policy. In certain communities, it’s perfectly acceptable that some of the members live apart as hermits most of the year, only breaking their meditative existence by checking in with the larger group on important holidays to share what they’ve been working on solo. In others, every hour has its appointed task, from prayer, to farming, to feeding people, to engaging in social activism.
The point is that everyone within a given community knows the basic guidelines, because at some point, they’ve been well-communicated. Beyond that, it is up to the individual to see whether they can happily live out their personal expression within the policy.
It’s a lot like retail can be, when done right. And it hinges on the question:
“Has culture been well-enough communicated to every employee so that he or she can act like the CEO of the company would in wide variety of circumstances?”
Or to put it another way, would Amazon owner Jeff Bezos be powerless to get me my vitamins?
The most accessible modern benchmark of good customer service — the online review — is what tells the public whether the CEO has “set the example.” Reviews tell whether time has been taken to acquaint every staffer with the business that employs them, preparing them to fit their own personal expression within the company’s vision of serving the public.
An employee who is able to recognize that an injured patron needs a seat while awaiting his prescription should be empowered to act immediately, knowing that the larger company supports treating people well. If poor training, burdensome chains of command, or failure to share brand culture are obstacles to common-sense personal initiative, the problem must be traced back to the CEO and corrected, starting from there.
And, of course, should a random staffer’s personal expression genuinely include an insurmountable disregard for other people, they can always be told it’s time to leave the monastery...
For marketing agencies, opportunity knocks
So your agency is auditing a valuable incoming client, and their negative reviews citing dirty premises, broken fixtures, food poisoning, slowness, rudeness, cluelessness, and lack of apparent concern make you say to yourself,
“Well, I was hoping we could clean up the bad data on the local business listings for this enterprise, but unless they clean up their customer service at 150 of their worst-rated locations, how much ROI are we really going to be able to deliver? What’s going on at these places?”
Let’s make no bones about this: Your honesty at this critical juncture could mean the difference between survival and closure for the brand.
You need to bring it home to the most senior level person you can reach in the organization that no amount of honest marketing can cover up poor customer service in the era of online reviews. If the brand has fallen to the level of the pharmacy I’ve cited, structural change is an absolute necessity. You can ask the tough questions, ask for an explanation of the bad reviews.
“But I’m just a digital marketer,” you may think. “I’m not in charge of whatever happens offline.”
Think again.
Headlines in retail land are horrid right now:
If you were a retail brand C-suite and were swallowing these predictions of doom with your daily breakfast, wouldn’t you be looking for inspiration from anyone with genuine insight? And if a marketing agency should make it their business to confront the truth while also being the bearer of some better news, wouldn’t you be ready to listen?
What is the truth? That poor reviews are symptoms smart doctors can use for diagnosis of structural problems.
What is the better news? The retail scenario is not nearly as dire as it may seem.
Why let hierarchy and traditional roles hold your agency back? Tesla wouldn’t. Why not roll up your sleeves and step into in-store? Organize and then translate the narrative negative reviews are telling about structural problems for the brand which have resulted in dangerously bad customer service. And then, be prepared to counter corporate inertia born of fear with some eye-opening statistics.
Print and share some good retail tidings
At Moz, we’re working with enterprises to get their basic location data into shape so that they are ready to win their share of the predicted $1.4 trillion in mobile-influenced local sales by 2021, and your agency can use these same numbers to combat indecision and apathy for your retail clients. Look at that second statistic again: 90% of purchases are still happening in physical stores. At Moz, we ask our customers if their data is ready for this. Your agency can ask its clients if their reputations are ready for this, if their employees have what they need to earn the brand’s piece of that 90% action. Great online data + great in-store service = table stakes for retail success.
While I won’t play down the unease that major brand retail closures is understandably causing, I hope I’ve given you the tools to fight the “retail disaster” narrative. 85% more mobile users are searching for things like “Where do I buy that reindeer moss vitamin D3?” than they were just 3 years ago. So long as retail staff is ready to deliver, I see no “apocalypse” here.
Investing time
So, your agency has put in the time to identify a reputation problem severe enough that it appears to be founded in structural deficiencies or policies. Perhaps you’ve used some ORM software to do review sentiment analysis to discover which of your client’s locations are hurting worst, or perhaps you’ve done an initial audit manually. You've communicated the bad news to the most senior-level person you can reach at the company, and you've also shared the statistics that make change seem very worthwhile, begging for a new commitment to in-store excellence. What happens next?
While there are going to be nuances specific to every brand, my bet is that the steps will look like this for most businesses:
C-suites need to invest time in creating a policy which a) abundantly communicates company culture, b) expresses trust in employee initiative, and c) dispenses with needless “chain of command” steps, while d) ensuring that every public facing staffer receives full and ongoing training. A recent study says 62% of new retail hires receive less than 10 hours of training. I’d call even these worrisome numbers optimistic. I worked at 5 retail jobs in my early youth. I’d estimate that I received no more than 1 hour of training at any of them. Because a chain of command can’t realistically be completely dispensed with in a large organization, store managers must then be allowed the time to communicate the culture, encourage employees to use common sense, define what “common sense” does and doesn’t look like to the company, and, finally, offer essential training. Employees at every level must be given the time to observe how happy or unhappy customers appear to be at their location, and they must be taught that their observations are of inestimable value to the brand. If an employee suggests a solution to a common consumer complaint, this should be recognized and rewarded. Finally, customers must be given the time to air their grievances at the time of service, in-person, with accessible, responsive staff. The word “corporate” need never come into most of these conversations unless a major claim is involved. Given that it may cost as much as 7x more to replace an unhappy customer than to keep an existing one happy, employees should be empowered to do business graciously and resolve complaints, in most cases, without escalation.
Benjamin Franklin may or may not have said that “time is money.” While the adage rings true in business, reviews have taught me the flip side — that a lack of time equals less money. Every negative review that cites helpless employees and poor service sounds to my marketing ears like a pocketful of silver dollars rolling down a drain.
The monk says good leaders make the time to communicate culture one-on-one.
Tesla says rules should change if they’re ridiculous.
Chairs should be offered to sick people… where common sense is applied.
Reviews can read like this:
And digital marketers have never known a time quite like this to have the ear of retail, maybe stepping beyond traditional boundaries into the fray of the real world. Maybe making a fundamental difference.
|
{
"pile_set_name": "OpenWebText2"
}
|
A political analyst says the United States is responsible for the genocide of the minority Rohingya Muslims in Myanmar, Press TV reports.
“The United States of America bears responsibility for this genocide, since the US has been rewarding the Myanmar regime with ever-closer political and economic ties during recent months of accelerating atrocities,” Kevin Barrett wrote in an article published on Press TV Website.ll
I am writing to every contact listed at Buddhanet.info's American
Buddhist Directory to ask:
Are you aware of the ongoing genocide in Myanmar (Burma) - a
genocide that is being committed in the name of Buddhism?
And did you know that the United States of America bears
responsibility for this genocide, since the US has been rewarding the
Myanmar regime with ever-closer political and economic ties during
recent months of accelerating atrocities?
As American Buddhists, you are in a position to help stop this
genocide, by pressuring the US and Myanmar governments as well as
international human rights organizations. Your visible participation
in the campaign to save the Rohingya people from extermination by
murderous Buddhist fanatics will not only help draw the world's
attention to this horrific situation, but also help restore the image
of Buddhism as a religion of compassion.
The facts about the genocide in Myanmar are not in dispute. The
fanatical Buddhist nationalists, who unfortunately represent a large
segment of the roughly 60 million Buddhists in Myanmar, admit that
they are trying to uproot and exterminate the roughly one million
Muslim Rohingya from land that the Rohingya have lived on for
centuries.
Here is what a typical genocidal Buddhist fanatic from Myanmar
wrote in a comment on a Wall
Street Journal article:
"Burma is Buddhist nation created for the 135 Tibeto-Burman
tribes. People do not get citizenship just because born there or
illegally lived there for centuries. Please do not interfere with the
law and internal affairs of Burma just as you do not like other
nations to poke their nose in your internal affairs."
"People do not get citizenship just because born there
or illegally lived there for centuries." This statement,
which aptly sums up the official policy of the Burmese regime, could
get the person who made it, and the government that follows it,
hanged for crimes against humanity. OBVIOUSLY being born in a modern
nation to a family that has been there for centuries automatically
confers citizenship. And OBVIOUSLY any modern nation that denies
citizenship to such people, burns their homes and communities, and
murders them en masse, with the aim of removing them from the nation
of their birth, is committing the internationally-recognized crime of
genocide.
In recent weeks, many thousands of homes, and more than 20
mosques, have been burned by murderous Buddhist mobs, backed by
national security forces, in the Arakan state of Myanmar. Estimates
of the number of Rohingya Muslims murdered, whether directly or by
drowning in the Naf river as they flee the killers, range from the
thousands to the tens of thousands. Every one of the more than 500
mosques in Arakan has been taken over by the genocidal regime's
security forces and shut down, and they are being demolished
one-by-one. (This happened during the holy month of Ramadan, when
Muslims are supposed to spend as much time as possible in a mosque.)
Muslims have been living in Burma since roughly 800 c.e. - that
is, nearly for as long as the religion of Islam has existed. And
Arakan has been a Muslim region, ruled by Muslim kings and/or
populated by Bengali Muslims, since 1430. The most notable population
increase of Muslims in Arakan took place in the 1600s. The idea that
the Rohingya people are somehow "recent immigrants" to the
region is clinically insane - a symptom of the larger insanity known
as nationalist fanaticism.
Both Buddhism and Islam are universalist religions: They proclaim
truths that are valid for all people, indeed for all of existence.
And the core truth that both religions proclaim is the primacy of
compassion. In Buddhism, a central feature of the Buddha nature is
compassion for all beings. If one were to choose a single hallmark of
a successful advanced practitioner of Buddhism, it would be a
highly-developed sense of compassion.
Whatever has happened to the Myanmar Buddhists' compassion for
their fellow citizens who happen to be born as Rohingyas?
Islam, too, views compassion as a central reality of creation.
Muslim theologians, like the more advanced Christian and Jewish
religious thinkers, view God as ineffable; but the primary and
overriding tangible characteristic of God in Islam (with the
proviso that no tangible characteristics fully express the reality of
the one ineffable God) is rahma, or compassion. The two
adjectives Muslims use the most to "describe" God are
ar-rahman ar-rahim, usually translated as "the merciful,
the compassionate." (The root of rahma and its cognates
derives from the word for "womb," suggesting that this
"compassion" has something in common with the nurturing,
all-embracing, unconditional love that mothers feel for their
children.)
Additionally, both Buddhism and Islam teach us to transcend or
even annihilate the (tribal) ego. Buddhism offers a set of teachings
that take its practitioners beyond the ego, which is the source of
the endless desire that is the cause of the pervasive suffering or
disappointment that characterizes ordinary human existence. Likewise,
Islam teaches its serious practitioners to annihilate the "ego
that desires evil" through absolute submission to God. Each
religion offers a very similar cure for the unhappiness of the
ordinary human condition.
The kind of chest-thumping egotistical nationalism that proclaims
"I am a Buddhist, my heroic nation is Buddhist, I am so much
better than those non-Buddhists that I must kill them or exile them"
is about as far from the compassionate teachings of the Buddha as it
is possible to get. Likewise, extremist Muslims who proclaim that
their narrow version of Islam is the only truth, and that everyone
who disagrees should be killed, are equally far from the universal,
all-compassionate message proclaimed by God through the Prophet
Muhammad (peace upon him).
Muslims and Buddhists ought to unite against ego-driven
nationalist fanaticism, which is an affront to both religious
traditions. A good starting point would be joining forces against the
genocide in Myanmar. Below are some suggestions for action.
Write and call Myanmar's
government contacts pointing out that every modern nation agrees
that anyone born inside a nation, whose parents and ancestors also
lived on that territory, is automatically a citizen of that nation
and must be protected by that nation's government.
|
{
"pile_set_name": "Pile-CC"
}
|
[Activating effect of bestim on the macrophages in experimental tuberculosis of varying severity].
The synthetic dipeptide bestim was tested for effects on the functional activity of peritoneal macrophages on 400 non-inbred albino mice while simulating generalized tuberculosis of varying severity (classical, acutely progressive, and slowly progressive). Bestim was shown to have a stimulating effect on the activity of macrophageal phagocytosis of yeast cell suspension. The agent was also found to exert a restorative effect on the absorptive and digestive functions of macrophages during their inhibition during infection and under the influence of long-term (more than a month) etiotropic therapy. Bestim showed an activating effect on the content of extracellular 5-nucleotidase during the classical and acutely progressive course of infection. It was shown to have a modulating effect on the macrophageal generation of superoxide radicals, by enhancing the inhibited HCT activity in acutely progressive infection and by reducing the elevated level of superoxide production in slowly progressive tuberculosis. In slowly progressive tuberculosis, the drug produced a stimulating effect on the adhesive activity of macrophages.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
So I went to a bar last night that Vinnie Paul used to frequent often when he was still alive and Alice Cooper's band was putting on a free show. Unfortunately the man himself wasn't there because they are doing a big show tonight downtown and last night was just a one-off private jam session which was awesome. Does anyone remember the guy that won that MTV VJ contest, Jesse Camp? He was there and he introduced the band and he even sung lost in America. It was a real hassle in pain in the ass getting to and from the bar but otherwise I had A blast!
That doesn't extend to abusing and/or openly fantasizing about violent acts towards other members. Not to mention - I've been a member here for over a decade, and I'm a moderator here...after what you said here
Quote:
I would love to spit roast those that gave me those infractions in the past.
being a clear attempt to poke the bear, and then
Quote:
Playing dumb, huh?
I'd be within my rights to show you the door again.
But I haven't - I'd suggest that if all you're here to do is express your dissatisfaction with events of the past and act in a disrespectful way to other members...find something new to post about and move forward, not back.
I remember you clearly.
Respect others - You'll get the same.
What you won't get is warnings, at least until you've worked your way a little bit out of the shitlist/s you're on right now.
I don't really want to have to talk about this any more.
Enjoy your time here. Have a look around. There's better shit to focus on.
|
{
"pile_set_name": "Pile-CC"
}
|
Genesis, this is not
It is one of the world’s classic stories.
A gigantic ark gets built with the help of a higher power. It is a huge, grandiose structure constructed out of wood that is perhaps larger than anything comparable in the world. Then heavy rains begin to fall, inundating the earth around it.
Distroscale
And that’s when the lawyers are called.
Genesis, this is not. Not in 2019, not in the United States.
Ark Encounter , the multimillion-dollar theme park and monument to fundamentalist Christianity whose centerpiece is a giant replica of Noah’s Ark, is suing its insurance carriers over rain-related damages on the property.
The company, Crosswater Canyon, is seeking to recoup what it says were $1 million worth of damages, as well as attorneys fees and costs, and an unspecified amount of punitive damages.
The irony has not been lost on local media organizations and observers on social media. Even the American Atheists took a shot on Twitter.
Story continues below This advertisement has not loaded yet, but your article continues below.
This particular chapter begins in 2017, when rains came to northern Kentucky. They did not fall for 40 days and 40 nights this time.
According to the National Weather Service, on-and-off precipitation throughout the year dropped 40-50 inches of rain that year on Williamstown, the town where the theme park is located – just a slight bump above average.
But a slope abutting an access road near the east side of the theme park’s ersatz ark began to fail in May, and eventually was subject to a “significant landslide” that took out a barrier along the road, according to the lawsuit.
“The Ark itself does not sit next to the damaged areas,” the theme park said Friday, taking issue with reports that attributed the damage to a flood. “The Ark was built on bedrock and was never in jeopardy.”
It reported the property damage to its insurance companies.
Photo by Aaron P. Bernstein/Getty Images
Engineers it hired recommended it replace the barrier with a retaining wall with drilled concrete shafts to prevent further damage, and the theme park also repaved and repaired portions of the road, the grading and added some drainage improvements. The total cost was about $1 million, the lawsuit says.
But the insurance companies denied claims for the improvements, saying that the policy had an exclusion for correcting design deficiencies or faulty workmanship.
The two sides went back and forth, according to the lawsuit and the insurance companies did pay a “very small portion” of Ark Encounter’s claim. But the organization says they have “breached their contractual obligations, acted in bad faith, and violated Kentucky law by failing to provide further coverage.”
Story continues below This advertisement has not loaded yet, but your article continues below.
The lawsuit was filed on Wednesday in district court in Kentucky. The Allied World Assurance Co. is named as a defendant, along with three other carriers, none of which responded to requests for comment.
The $120 million Ark Encounter, where adult tickets cost $48, was completed in 2016, with a zoo, zip lines, and a restaurant in addition to its five-storey high replica ark. It was the brainchild of Ken Ham and his ministry, Answers in Genesis, which also created the Creation Museum. Ham, is a Christian fundamentalist and creationist who argues that the Bible is a historical narrative that is meant to be taken literally.
Photo by Bloomberg photo by Luke Sharrett
He believes that dinosaurs lived alongside humans and that the biblical flood created the Grand Canyon. And he maintains that Noah labored seven decades to construct his vessel and was 600 years old when the storm surged.
He did not respond to a request for comment sent to Ark Encounter spokeswoman Melany Ethridge.
“The lawsuit speaks for itself,” read a statement she distributed on behalf of the park. “Ark Encounter guests have been unaffected by the work being done at the access road. Hours of operation were never affected.”
The theme park was met with no small amount of controversy when it opened, focused mainly on its sources of funding, including the $62 million in junk bonds that were floated by the town of Williamstown.
The ACLU and other groups charged that Answers in Genesis should have not been eligible for state and local subsidies because of its discriminatory hiring practices.
As a condition of employment, the museum and ark staff of 900, including 350 seasonal workers, must sign a statement of faith rejecting evolution and declaring that they regularly attend church and view homosexuality as a sin.
|
{
"pile_set_name": "OpenWebText2"
}
|
[ Thanks to Steven J. Vaughan-Nichols for this link. ]
"I say yes.. My Samsung Series 5 Chromebook, which I've been using for months now, is the perfect grab and go laptop. It's weights just over three-pounds, the battery lasts for about ten hours, and the lightweight Linux desktop with a Chrome Web browser interface is all I need for work out of the office.
"That said, the first generation of Chrome OS had its problems. On the other hand, since then Google has made numerous significant improvements to Chrome OS and almost every week sees new improvement to Chrome OS, the Chrome Web browser, and Google's family of cloud-based applications that Chromebooks use in lieu of traditional desktop apps."
|
{
"pile_set_name": "OpenWebText2"
}
|
{#sp1 .447}
{#sp2 .448}
|
{
"pile_set_name": "PubMed Central"
}
|
Q:
CSS Selector - first element with Class in a group of same class
I have a list of elements under the same container. some of them has class A, other has class B.
I'm looking for a CSS selector to select the first A element in a group of A and the first B element in a group of B.
Example:
<style>
.A { background: red; }
.B { background: blue; }
</style>
<ul class="list">
<li class="item A">AAA</li><!-- I want to select this -->
<li class="item A">aaa</li>
<li class="item A">aaa</li>
<li class="item B">BBB</li><!-- I want to select this -->
<li class="item B">bbb</li>
<li class="item B">bbb</li>
<li class="item A">AAA</li><!-- I want to select this -->
<li class="item A">aaa</li>
<li class="item A">aaa</li>
</ul>
Notice that using the pseudo selector :first-of-type will not work, becuase it will select the first element of type without group relation.
Also trying to use :first-child will not work ether, since it will select only the first A element, and will not select the first B element, also it will not select the first A element in the 2nd group.
Note
I don't believe this is a duplicate of "CSS selector for first element with class".
A:
To select first child of the ul element, you can use :first-child pseudo-class.
To select the first element in each group, you can use adjacent sibling selector.
.A + .B will select any element with class B that immediately follows an element with class A. Similarly .B + .A will select any element with class A that immediately follows an element with class B
.A { background: red; }
.B { background: blue; }
.A:first-child,
.B + .A,
.A + .B {
background: yellow;
}
<ul class="list">
<li class="item A">AAA</li><!-- I want to select this -->
<li class="item A">AAA</li>
<li class="item A">AAA</li>
<li class="item B">BBB</li><!-- I want to select this -->
<li class="item B">BBB</li>
<li class="item B">BBB</li>
<li class="item A">AAA</li><!-- I want to select this -->
<li class="item A">AAA</li>
<li class="item A">AAA</li>
</ul>
|
{
"pile_set_name": "StackExchange"
}
|
Knowledge and self-reported use of universal precautions in a university teaching hospital.
This article reports a study of knowledge and self-reported use of universal precautions by professionals (individuals on the medical staff, in nursing service, in the laboratory, and in the social service, pastoral care, and respiratory care departments) and nonprofessionals (individuals working in the dietary and environmental services departments) in a university teaching hospital. Nonprofessional workers were found to be less knowledgeable about universal precautions than were professional workers. Appropriate use of universal precautions significantly correlated with a worker's perceived risk of exposure (positively correlated), the belief that HIV is easy to "catch" (negatively correlated), and the extent of the worker's involvement with HIV-infected patients (positively correlated). Use of universal precautions was not found to be significantly correlated with belief in universal precautions as a reasonable way to protect oneself from contracting HIV in the workplace. Recommendations are made for increasing appropriate use of universal precautions.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
World Health Organization: Fight Against Malaria Slowing Down
LONDON (AP) - A report from the World Health Organization says the fight against malaria is slowing down amid a dramatic drop in efforts to reverse the epidemic.
Two years ago, about 145 million bed nets were given out across Africa to protect people against the mosquitoes that spread the killer disease. Last year, that number fell to about 66 million. The number of homes in Africa sprayed with pesticides has also stalled, as have attempts to treat pregnant women, one of the high-risk groups.
While the malaria death rate has fallen by a quarter since 2000, officials say further improvements are in jeopardy.
WHO Director-General Dr. Margaret Chan says "urgency and determination" are needed to keep making progress toward the goal of eliminating malaria by the end of 2015.
WHO estimates there were about 219 million malaria cases and 660,000 deaths in 2010.
Malaria causes symptoms including fever, chills, and vomiting and can kill if not treated early. It mainly strikes children under 5, mostly in Africa.
Copyright 2015 Associated Press. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.
Online Public Information File
Viewers with disabilities can get assistance accessing this station's FCC Public Inspection File by contacting the station with the information listed below. Questions or concerns relating to the accessibility of the FCC's online public file system should be directed to the FCC at 888-225-5322, 888-835-5322 (TTY), or [email protected].
|
{
"pile_set_name": "Pile-CC"
}
|
The specific aims of this research are to evaluate the quality of published clinical trials, suggest improvements, and combine data where appropriate to enable conclusions to be drawn from multiple, frequently undersized therapeutic and diagnostic trials. The impact of this research on practicing physicians is measured by assaying review articles, sending out questionnaires and reviewing hospital charts. Reading habits of superior and inferior physicians (as determined by peer review) are to be documented. Throughout these studies care is taken to minimize bias by selective photocopying and by working in duplicate to minimize errors. Medical students gather the data under supervision and present and write the papers. The biostatisticians are working on methods of analyzing combined data and improving the handling of survival data. A computerized file of randomized control trials is being developed and a data base of multiple trials designed so that essential combined and individual data can be made available on line to practicing physicians. It is anticipated that these efforts will go a long way towards facilitating the evaluation and transmission of ever increasing amounts of therapeutic and diagnostic studies to students and deliverers of health care.
|
{
"pile_set_name": "NIH ExPorter"
}
|
export { default as SurveyList } from './SurveyList';
export { default as SurveyQuestionAdd } from './SurveyQuestionAdd';
export { default as SurveyQuestionEdit } from './SurveyQuestionEdit';
export { default as SurveyPreviewModal } from './SurveyPreviewModal';
|
{
"pile_set_name": "Github"
}
|
Synthesis and in vitro anticancer activity of octahedral platinum(IV) complexes with cyclohexyl-functionalized ethylenediamine-N,N'-diacetate-type ligands.
The present study describes the synthesis and anticancer activity of novel octahedral Pt(IV) complexes with cyclohexyl functionalized ethylenediamine-N,N'-diacetate-type ligands. Molecular mechanics calculations and density functional theory analysis revealed that s-cis is the preferred geometry of these Pt(IV) complexes with tetradentate-coordinated (S,S)-ethylenediamine-N,N'-di-2-(3-cyclohexyl)propanoate. The viability of cancer cell lines (U251 human glioma, C6 rat glioma, L929 mouse fibrosarcoma, and B16 human melanoma) was assessed by measuring mitochondrial dehydrogenase activity and lactate dehydrogenase release. Cell-cycle distribution, oxidative stress, caspase activation, and induction of autophagy were analyzed by flow cytometry using appropriate fluorescent reporter dyes. The cytotoxic activity of novel Pt(IV) complexes against various cancer cell lines (IC(50) range: 1.9-8.7 microM) was higher than that of cisplatin (IC(50) range: 10.9-67.0 microM) and proceeded through completely different mechanisms. Cisplatin induced caspase-dependent apoptosis associated with the cytoprotective autophagic response. In contrast, the new Pt(IV) complexes caused rapid, caspase-independent, oxidative stress-mediated non-apoptotic cell death characterized by massive cytoplasmic vacuolization, cell membrane damage, and the absence of protective autophagy.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Laparoscopic cholecystectomy in children with chronic hemolytic anemia. Is the outcome related to the timing of the procedure?
The aim of this study was to evaluate whether the outcome in children with chronic hemolytic anemia (CHA) and cholelithiasis undergoing laparoscopic cholecystectomy (LC) is related to the operation timing. From June 1995 to December 2004, 46 children with CHA were referred to our division of surgery for cholelithiasis. All 46 children were asymptomatic at the time of the first visit, and an elective LC was proposed to all of them before the onset of symptoms. The operation was accepted in the period of study by 24 children and refused by 22. The patients were divided into three groups (group A, asymptomatic; group B, symptomatic; and group C, emergency admitted) depending on clinical presentation and operation timing, and the respective outcomes were compared. Elective LC in asymptomatic children (group A) is safe with no major complications reported. In children who refused surgery (groups B and C), we observed four sickle cell crises, four acute cholecystitis, and two choledocholithiasis, and all these complications were related to waiting. Two sickle cell crises occurred in symptomatic children waiting for surgery during biliary colic. The risk of emergency admission in children with cholelithiasis and CHA awaiting surgery was found to be high: 28% of the children admitted in emergency after a mean of 32 months (range, 22-36). Morbidity rate and postoperative stay increased when children with hemoglobinopathies underwent emergency LC. Elective LC should be the gold standard in children with CHA and asymptomatic cholelithiasis in order to prevent the potential complications of cholecystitis and choledocholithiasis, which lead to major risks, discomfort, and longer hospital stay.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
A PRECIOUS, mouldering pleasure ’t is
To meet an antique book,
In just the dress his century wore;
A privilege, I think,
His venerable hand to take, 5
And warming in our own,
A passage back, or two, to make
To times when he was young.
His quaint opinions to inspect,
His knowledge to unfold 10
On what concerns our mutual mind,
The literature of old;
What interested scholars most,
What competitions ran
When Plato was a certainty, 15
And Sophocles a man;
When Sappho was a living girl,
And Beatrice wore
The gown that Dante deified.
Facts, centuries before, 20
He traverses familiar,
As one should come to town
And tell you all your dreams were true:
He lived where dreams were born.
His presence is enchantment, 25
You beg him not to go;
Old volumes shake their vellum heads
|
{
"pile_set_name": "OpenWebText2"
}
|
Passive cavitation mapping with temporal sparsity constraint.
The spatial resolution of cavitation maps generated from passive recordings of cavitation emissions is compromised by the bandlimited nature of the recordings. Deconvolution based on the assumption that cavitation consists of a sparse series of discrete events allows the recovery of frequency components that are not only outside the frequency band of the receivers, but may also have been attenuated by the medium before being detectable. In the current work, two sparse deconvolution techniques, matching pursuit and basis pursuit, were applied to simulated and experimental cavitation recordings before they were beamformed to provide passive maps of cavitation activity. Matching pursuit was shown to reduce the maximal diameter of the point spread function by almost a third, at the cost of greater susceptibility to inter-source interference. In contrast, although basis pursuit causes an almost 20% increase in the maximal diameter of the point spread function, its application to experimental data appears to enhance the ability of passive mapping to resolve multiple sources.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Exogenous leptin inhibits the secretion of pancreatic juice via a duodenal CCK1-vagal-dependent mechanism in anaesthetized rats.
Leptin originally described as product of the ob gene has been shown to be expressed in various tissues including the gastrointestinal tract. In this study, we investigated the influence of leptin on the secretion of pancreatic juice in biliary-pancreatic duct cannulated anaesthetised rats and in dispersed rat pancreatic acini in vitro. Exogenous leptin was given in boluses intravenously with or without CCK-8 (12 pmol kg(-1) body weight) in the presence or absence pharmacological CCK(1) receptor blockade, cervical vagotomy, and capsaicin pre-treatment. Administration of leptin (0.1, 1 and 10 microg kg(-1) body weight) did not affect the volume of bile and pancreatic juice while the protein and trypsin outputs were reduced in a dose-dependent manner. In the rats, leptin inhibited CCK-8 stimulated protein and trypsin outputs stronger than the basal pancreatic secretion. The inhibition by leptin was abolished by the pharmacological CCK(1) receptor blockade, cervical vagotomy, and capsaicin pre-treatment. In contrast, leptin did not affect basal and CCK-8-stimulated amylase release from the dispersed rat pancreatic acini in vitro. In conclusion, the results of the present study suggest that leptin does not act directly on the rat pancreatic acinar cells but inhibits the secretion of pancreatic enzymes acting indirectly via a neurohormonal CCK-vagal-dependent mechanism.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Q:
How can I dealloc manually UIViewController from another UIViewController in UINavigationController without import the first UIViewController class?
I've a navigationController. The first viewcontroller is a kind of class FirstViewController. When I tap a button in FirstViewController, it push in navigationController the second viewController that is a kind of class SecondViewController. When I tap a button in SecondViewController, I'd like to dealloc the FirstViewController (previously saved in navigationController) so as to start again as if it was the first time that I open the FirstViewController when I tap back button in navigationItem of SecondViewController. Here the code of the method called when I tap the button in secondViewController:
NSArray * navigationPath = [self.navigationController viewControllers];
UIViewController *previousVC = [navigationPath objectAtIndex:[navigationPath count]-2];
[previousVC performSelector:@selector(viewDidUnload)];
It doesn't dealloc the FirstViewController. There is a way to do it?
A:
Create a new first view controller
FirstViewController *first = [[FirstViewController alloc]init];
Reset the navigation stack
[self.navigationController setViewControllers:@[first, self] animated:NO];
Navigate to first
[self.navigationController popViewController:YES];
|
{
"pile_set_name": "StackExchange"
}
|
1. Field of Invention
This invention relates to hosts containing constructs with genes from the starch pathway. More specifically the present invention relates to bacterial hosts that form plant like starches. Additionally the present invention relates to plant hosts that have genes from the starch pathway. The invention further relates to the starches produced by said hosts.
2. Description of Prior Art
The starch using industry includes diverse industries such as candy makers, makers of adhesives and paints, gravy makers, paper producers, etc. Since the demand for starch, (which is formed of amylose and amylopectin), has been dramatically increasing for specialized food and industrial uses, efforts have been undertaken to tailor the quantity and quality of starch for specific food and industrial uses.
This industry has over time looked for a number of different starches having, high viscosity, lower viscosity, higher gelling strength and lower gelling strength, different boiling points etc. Each starch tailored for a number of uses. The industry has utilized mutant starches that have less amylopectin and mutant starches with more amylose for tailored specifications. For example the increased amylose starch has been used in the gelled candy making area. And the industry has used the increased amylopectin starches formed by mutants such as wx and wx su2 containing little amylose and mostly amylopectin for thicken foods like pudding, pies, gravies, frozen foods and batters, stews, canned foods and baby food. Additionally the mutant starches of different types have usefulness as adhesives and as sizing.
The other method used to address the industry needs for tailored starch is the use of chemical modification of the starch. Chemical derivation of the starch are produced by chemically reacting the starch with the monofunctional reagents to introduce the substituents such as phosphate, acetate, succinate groups to stabilize the starch. Unfortunately, these types of starches can be subjected to government regulation and additionally have less acceptance generally due to the added cost of the starch.
Starch is the major form in which carbohydrates are stored in biological systems. Plant starch in chloroplasts is transitory and storage starch accumulates in storage organs of many plant. Starch can be found in all organs of most higher plants including leaves stems and roots and fruits and embryo and endosperm. In addition to higher plants starch similar polysaccharide (glycogen) has been found in bacteria. Many bacteria produce a reserve polysaccharide similar to the glycogen found in animals.
Storage polysaccharide has been classified as being in two groups, group one has storage in the cystol of the cell and the second group within the plastid. Escherichia coli produces a polysaccharide within the cytosol. Starch producing plants typically store starch in the plastids. Typical starch bearing plants include cassava, potato, corn, peas, rice, wheat, and barley. The main starch storing tissue of corn, rice wheat and barley and oats, the cereal grains, is the endosperm.
Starches are also classified by the plant source, for example cereal starches are from cereal grains such as maize, rice, wheat, barley, oats and sorghum; tuber and root starch are from potatoes and yams and cassava.
The pathway of starch synthesis is not well understood. Generally, as noted above starch from plants, consists of two major components: amylose and amylopectin. These intertwine in the starch granule of the plants. Amylose is a linear polymer of alpha 1-4 bonded anhydroglucose units while amylopectin is a branched polymer comprised of linear chains alpha 1-4 linked anhydroglucose units with branches resulting from alpha 1-6 linkages between the linear chains. It has been known for sometime that mutant genes in starch bearing plants can be expressed and that the properties of the starch can be altered. The proportion of the two components and their structures in the mutant primarily determine the physical-chemical properties of the starch.
Thus the lack of a clear understanding of the starch synthesis pathway and the difficulty of employing mutants limited the industry to the use of existing and producible mutant starches (cereals containing mutant starch can show a tremendous yield penalty in field environments) or to the chemical modification that could be made to the starch. In the last decade the industry has been studying the effects of certain starch genes in plants and bacteria in an attempt to more clearly understand starch synthesis. Since the late 80's it has been possible to transform plants and bacteria to contain isolated genes. In response to this the industry has transformed potatoes with a bacterial gene GS and with starch soluble synthase III gene in the antisense (forming a mutant). As part of these potato starch experiments bacteria has been transformed with certain potato starch genes. For example the SSSIII gene from potato was transformed into E. coli deficient in the glgA gene. The effect of glgC and branching enzyme I and II in combination in a mutant E. coli has also been studied and glycogen like product was reported. The starch industry that is commercial does not have a particular interest in the production of glycogen which is the polysaccharide produced by bacteria and animals (the health care industry may have some such interest). The industry has thus not yet been able to generate tailored starches at reasonable prices through plant gene transformation. There remains a need for the industry to find new starches that are useful due to their changed characteristics such as lower viscosity and new starches that are useful because of their higher viscosity and new methods of producing such starches.
Glycogen synthesis in E. coli and starch synthesis in higher plants have similar pathways involving ADPGlc pyrophosphorylase, starch synthase (SS), or glycogen synthase (GS), and branching enzyme (BE). It has been suggested that ADPGlc pyrophosphorylase plays a pivotal role in regulating the amount of starch synthesized, while starch synthase and starch branching enzyme primarily determine the starch structure. Multiple forms of SBE and SS have been identified in many plants including maize, rice, pea and potato. In addition to the waxy gene coding for granule bound starch synthase (GBSS), three genes coding for the other forms of SS have been isolated from maize endosperm. Maize is the only cereal crop from which the genes coding for the five forms of SS have been isolated. Clearly a number of these sequences are published and known to those of skill in the art. Genes coding for maize SBE have also been cloned and characterized. Previous reports have demonstrated that maize SBEI has a higher rate of branching amylose than SBEII and preferentially transfers longer chains, while SBEII shows a higher rate of branching amylopectin and preferentially transfers shorter chains. In comparison with SBE, less is known about the specificities and functions of multiple forms of SS. In Waxy maize, which lacks GBSS, only amylopectin is synthesized and amylose is missing. Therefore, it is generally accepted that GBSS, encoded by waxy gene, is primarily responsible for the synthesis of amylose. Study of waxy mutation in Chlamedomonas reinhardtii has suggested that GBSS is also involved in amylopectin synthesis. Although it has been reported that Chlamedomonas reinhardtii SSII controls the synthesis of intermediate size glucans of amylopectin in Chlamedomonas, direct evidence for the functions of SS in higher plants is still missing. Antisense technology has been used to study the functions of SS in potato, however, the results are inconclusive.
In an article written by Hanping Guan et al., entitled “A Maize Branching Enzyme Catalyzes Synthesis of Glycogen-like Polysaccharide in glg B-deficient Escherichia coli”, published in Proc. Natl. Acad. Sci. USA, Vol. 92, pp. 964-967, February 1995 Plant Biology a specific glycogen like polysaccharide from a transformed E. coli was reported. This article taught the transformation of an E. coli bacteria with maize BEI and BEII. These enzymes were transformed into two E. coli hosts. One of the bacterial hosts was a wild type and the other was a mutant. The mutation to the bacteria was the reduction of the activity of glycogen BE in the AC71(glgB-) so that the mutant was essentially free of BE activity. The paper analyzed the debranched alpha-glucan isolated from the four different transformants. The first host was E. coli containing glgB and the second host was the AC71 without any transformed genes then AC71 with maize BEII, and then AC71 with maize BEI, then AC71 with maize BEI and BEII. The resultant polysaccharide products were analyzed by HPLC, by chain length and relative peak area and by mole distribution of chains. The study led to the understanding that BEII could play a role in synthesis of the short chains of amylopectin and BEI could be involved with the longer chains of amylopectin. The paper also noted that the mutant host AC71 produced more chains with chain length of 6 then did the wild type E. coli. The paper also noted that the maize BE and the GS of the host did not produce amylopectin like polysaccharides. The article suggested that the concerted action of GS with different BE's could play an important role in determining the final structure of the polysaccharide synthesized. The article by Guan ends by suggesting that his study had established the basis for studying the concerted actions of BE and SS in a bacterial model system.
The expression of E. coli GS (glycogen synthases) in potatoes showed a large incidence of highly branched starch. This result was published in an article in Plant Physiol. 104, 1159-1166 by Shewaker et al. This potato does not appear to be of much commercial use at this time.
The industry still needs the option of producing plant like starches in a fermentation process from bacteria and thus without the necessity of breeding and growing environment sensitive plants; and, the option of producing plants that generate the specific tailored starch through a plant host. And the industry needs altered and new starches that are cereal like starches or root and tuber like starches in large quantities and inexpensively thus avoiding having to use chemical modification of starch. The industry needs a host that can be readily transformed to supply different starches tailored to the industry's need. Specifically the industry needs a host that supplies various different starches including those not capable of being made in plants or bacteria presently.
Objects and Advantages
Accordingly, several objects and advantages of the invention are to produce plant like starch through the process of fermentation.
Additional objects and advantages are the production of new starches in plants.
Still further objects and advantages will become apparent from a consideration of the ensuing description and accompanying drawings.
Another object of the present invention is the synthesis of polysaccharides including amylose, amylopectin in E. coli, and/or fungal and yeast by plant starch synthesizing enzymes including SS, SBE, bacterial branching enzyme, glycogen synthase and other enzymes in other living organism.
Yet another object of my invention is using each or combination of these enzymes or modified enzymes studied in this patent to produce or to improve polysaccharides in any living organism including starch synthesis in plants.
|
{
"pile_set_name": "USPTO Backgrounds"
}
|
Four-player madness! You can’t spell Nippon Marathon without ‘P’, ‘A’, ‘R’, ‘T’ and *checks letters* ‘I’, we suppose… Solo play is supported, but grab up to four pads and enjoy the supreme silliness o... Read More
|
{
"pile_set_name": "OpenWebText2"
}
|
Unix Windowing Terminal System Blit, 1982, Bell Labs - dryicerx
http://www.youtube.com/watch?v=waTL1abCm9I
======
daeken
I've spent the last two days implementing VGA and basic graphics support in my
OS, and it's great to watch this and see something I know I can implement in a
short period of time with what I've already built. Good for morale if nothing
else.
------
mahmud
_Unix compilers are slow, so to entertain myself while I am waiting, I can
play Asteroids!_
Good lord. The whole video had an underlying tone of _fun_.
I have read this repeatedly and heard it in person from several older
engineers: "unix is a toy", "unix is for playing video games and writing term
papers", "unix is for usenet", etc.
During that time, if I was a buttoned down government or corporate programmer
with IBM and Honeywell training, jockeying Cobol, assembly and Fortran, I too
would eschew this newfangled game-playing computer.
[Edit: I was gonna say Dan Ingalls invented bitblit operation that made 2d
graphics possible, then I saw he had a hand in the production of the video as
well ;-) ]
------
dryicerx
_When I want to compile a program, I don't have to exit the program, I just go
to another layer and type make_
Although my current setup for editing and compiling is still eerily identical
to that.
|
{
"pile_set_name": "HackerNews"
}
|
Image caption William Hague reportedly described top Tories as "staunch Atlanticists"
Files newly released by the Wikileaks website highlight what is described as the UK's "paranoia" about its so-called special relationship with the US.
In one cable, a senior US diplomat describes "excessive UK speculation" after Barack Obama became president.
"This over-reading would often be humorous, if it were not so corrosive," it says, according to the Guardian.
The cables detail efforts by leading Tories, who are now in government, to stress their pro-US credentials.
A 2008 cable written by US deputy chief of mission Richard LeBaron describes a meeting with William Hague, then a Conservative frontbencher and now foreign secretary.
When asked at the meeting whether the relationship was "still special", Mr Hague is said to have replied: "We want a pro-American regime. We need it. The world needs it."
The messages are among more than 250,000 US cables obtained by Wikileaks and being published in stages by the whistle-blowing website.
Details are also being published in the Guardian and several other papers around the world that gained an advance look at the material.
'Unparalleled commitment'
"The British ask, is our special relationship still special in Washington?" is the heading of a message from Mr LeBaron in February 2009.
Image caption The return of a Churchill bust on loan to the US caused British anxiety
He writes: "More than one [UK] senior official asked embassy officers whether President Obama meant to send a signal in his inaugural address about US-UK relations by quoting Washington during the revolutionary war, while the removal of the Churchill bust from the Oval office consumed much UK newsprint.
"This period of excessive UK speculation about the relationship is more paranoid than usual.
"This over-reading would often be humorous, if it were not so corrosive."
Mr LeBaron goes on to advise resisting the temptation to keep the UK "off balance about its current standing with us" in order to foster greater assistance.
Analysis What comes over in these cables is a pervasive British sense of insecurity about the relationship with Washington. This was heightened by the way some of the early actions of President Barack Obama were interpreted. The British press and commentators, a senior US official reports back to Washington, were alarmed that Mr Obama referred to America's Revolutionary War leader George Washington in his inaugural address. Reports that Mr Obama had removed a bust of Churchill from the Oval Office prompted additional concerns. If reassurance were needed about the abiding importance of the close ties between London and Washington, then the US official sums it up in this way. He says "the UK's commitment ... remains unparalleled. A UK public confident that the US government values those contributions and our relationship matters to US national security." Nonetheless a fundamental question remains. As US foreign policy looks more towards Asia and the Pacific, how can Europe let alone just Britain's salience in Washington's outlook be preserved?
"The UK's commitment of resources - financial, military, diplomatic - in support of US global priorities remains unparalleled; a UK public confident that the [US] values those contributions and our relationship, matters to US national security."
Another cable from Mr LeBaron records Mr Hague describing Tory colleagues David Cameron, George Osborne and himself as "'children of Thatcher' and staunch Atlanticists".
"For his part, said Hague, he has a sister who is American, spends his own vacations in America and, like many similar to him, considers America the 'other country to turn to'," the cable says.
At a meeting with US ambassador Louis Susman just months before the UK general election, Liam Fox reportedly "affirmed his desire to work closely with the US if the Conservative Party wins power".
Mr Fox, now defence secretary, said the Tories would "follow a much more pro-American profile in procurement" of military equipment, according to Mr Susman's cable.
The message goes on: "Fox asserted that some within the Conservative party are less enthusiastic, asserting that 'we're supposed to be partners with, not supplicants to, the United States'.
"Fox said he rebuffed these assertions, and he welcomed the ambassador's reassurance that senior US leaders value the UK as an equal partner."
|
{
"pile_set_name": "OpenWebText2"
}
|
Q:
If $\cos 17x = f(\cos x)$, then show that $\sin 17 x=f(\sin x)$
If $f$ denotes the function which gives $\cos(17x)$ in terms of $\cos x$, that is $\cos(17 x) = f (\cos x)$, then, prove that it is the same function $f$ which gives $\sin(17x)$ in terms of $\sin x$. Generalize this result.
I am really unable to understand how I should go about this question.
Thanks a lot.
A:
$\displaystyle \cos 17x = f(\cos x) \\ \displaystyle\cos 17 (\frac{\pi}2 - x) = f(\cos (\frac{\pi}2 - x)) \\ \displaystyle\cos (\frac{17\pi}{2} - 17x) =f(\sin x) \\\displaystyle \cos (8\pi + \frac{\pi}{2} - 17x) = f(\sin x) \\ \displaystyle\cos(\frac{\pi}2 - 17x) = f(\sin x) \\\displaystyle \sin 17x = f(\sin x) $
The generalisation is that $\displaystyle \cos nx = f(\cos x) \iff \sin nx = f(\sin x)$ holds for all $\displaystyle n = 4k+1, k \in \mathbb{Z}$
|
{
"pile_set_name": "StackExchange"
}
|
Got a great wee magnetic needle minder for to keep my needles with my current project, and not only that it has the saying from the Harry Potter series that makes the Marauders' Map work.
Excellent wee present.
Edit: got a parcel in mail today (5th Aug) and opened up to a box full of goodies - bar of chocolate for to recover from a minor dementor attack, a sorting cauldron (instead of sorting hat), a test tube to keep wing of bat but is full of (hopefully Bertie Botts) jelly beans, a small red bag full with more beans, a small jar of magic dust (not sure what proper use of it is), an A5 sheet with spells written on it (may frame) and a scroll with a long rhyming spell written on it, and a one way Hogwarts Express ticket.
|
{
"pile_set_name": "OpenWebText2"
}
|
Q:
How to generate one geojson file from several rows in Postgres/PostGIS?
I have a table with geospatial objects, in particular, polygons. I know that I can generate from one row a geoJSON string using ST_AsGeoJSON(http://www.postgis.org/docs/ST_AsGeoJSON.html). But I would like to generate one geoJSON from all the table. Is this possible? How can I do it?
Thanks in advance
A:
Already make a comment but in addition if you choose OGR2OGR, the way to export to GeoJSON is the following:
If you want to keep original SRS
ogr2ogr -f "GeoJSON" yourdataoutputname.geojson PG:"host=localhost user=username dbname=yourdatabasename password=yourpassword" "yourpostgistable"
If you want to convert to default SRS for GeoJSON (EPSG 4326)
ogr2ogr -f "GeoJSON" yourdataoutputname.geojson -t_srs EPSG:4326 PG:"host=localhost user=username dbname=yourdatabasename password=yourpassword" "yourpostgistable"
|
{
"pile_set_name": "StackExchange"
}
|
RecruitGibraltar are currently helping our client who is a leading online based company who has a great opportunity for an Online Experience Project Manager. You will be required to co ordinate a number of high profile online projects including timelines and budgets. You will need to tie in a road map for design, content management, front end development and work cooperatively with key team members, internal and external clients. You will be expected to participate in team meetings, discussions and brainstorming sessions to create ongoing project timelines and monitor internal and external milestones.
The role:
? Managing design projects to improve website usability- briefing up to four different designers, freelancers and agencies at once but bringing in projects on time and in budget
? Overseeing the front end design of a new website launch- managing designers and liaising with the business- IT, marketing, the CEO and other stakeholders to deliver an optimum customer experience
? Must have a good eye for detail, and have worked on web projects or have briefed web design/ usability
? The candidate does not need to have design skills but must be able to manage design in the sense of briefing and owning projects
? The candidate must be able to work to tight deadlines
? The candidate must understand the principles of web usability
? The candidate must be able to influence at senior level
? A key part of the role is also manage smaller web projects- for example creating new payment screens and delivering them to our IT department, ensuring they are compliant and user friendly and designed within budget.
Salary 30k - 35k depending on experience.
To apply please send your CV to [email protected] and state which role you are interested in.
|
{
"pile_set_name": "Pile-CC"
}
|
INTRODUCTION
============
Pectus excavatum (PE) or funnel chest is a congenital chest wall deformity with an incidence of 1 in 400 live births and a maleto-female ratio of 4:1 \[[@b1-aps-2019-00318],[@b2-aps-2019-00318]\]. In female patients, the deformity commonly leads to a distortion of the female breast, resulting in so-called breast strabism with a diminished intermamillary distance \[[@b3-aps-2019-00318]\]. Females with PE tend to complain about the disturbing aesthetic appearance of their breasts and not about the sternal depression, which may be partly concealed by their breast tissues. Funnel chest resulting in a severe breast deformity may stigmatize a young woman, causing embarrassment, social anxiety, and other psychological disorders \[[@b4-aps-2019-00318]\]. Surgical correction of the deformity has been shown to improve patients' quality of life and to have positive influence on self-perception \[[@b5-aps-2019-00318],[@b6-aps-2019-00318]\].
Trying to correct the deformity by augmenting the breast with silicone implants may aggravate the strabism of the breasts by making the sternal inclination more prominent, which leads to unsatisfactory aesthetic results \[[@b3-aps-2019-00318]\]. Instead, as we have demonstrated in previous publications, remodeling the anterior thoracic wall with retrosternal metal bar implantation by a minimally invasive repair procedure (MIRPE) \[[@b7-aps-2019-00318]\] or by a modified approach with a semi-open technique (minor open videoendoscopically assisted repair of pectus excavatum; MOVARPE) \[[@b8-aps-2019-00318]\] corrects the underlying deformity and leads to a marked improvement in the appearance of the breasts \[[@b3-aps-2019-00318]\].
Aesthetic correction remains a major indication for PE repair and is the most difficult parameter to measure postoperatively. The patient's satisfaction with a procedure and the surgeon's evaluation of the postoperative results can differ \[[@b9-aps-2019-00318],[@b10-aps-2019-00318]\]. The aim of this study was to investigate the association between female patients' personal outcome evaluations and the assessments made by medical professionals. Standardized photographs of patients' preoperative and postoperative aesthetic appearance were analyzed by patients and by medical professionals. Differences in the evaluation based on evaluators' sex or degree of medical education were determined. Furthermore, we assessed whether the surgical intervention had an impact on the psychosocial well-being of the female participants. To our knowledge, no publications have yet investigated quality of life, satisfaction, body image, and aesthetic results before and after bar removal with an exclusive focus on female PE patients.
METHODS
=======
Study design
------------
The study design was prospective and approved by the local ethics committee (approval No. UN4960). All patients signed a written informed consent form and the data were analyzed anonymously. The study group included female patients who underwent correction of PE utilizing the MIRPE technique \[[@b7-aps-2019-00318]\] or the MOVARPE \[[@b8-aps-2019-00318]\] technique between August 2002 and February 2015 and had the pectus bar removed. Exclusion criteria were male patients, patients who underwent corrections using other techniques, or female patients who still had the pectus bar *in situ*. The patients underwent a clinical examination and standardized photographic documentation before and after correction.
All patients completed two questionnaires regarding their preoperative and postoperative conditions and their psychological well-being. We used specifically constructed questionnaires focused on women to evaluate their quality of life, satisfaction, and body image. The two-step Nuss Questionnaire (Pectus Excavatum Evaluation Questionnaire), which assesses the effects of the Nuss procedure on physical and psychosocial functioning, was developed and has been applied for pediatric populations. Modifications of this instrument have been made for adult male patients \[[@b11-aps-2019-00318]\], but it has not yet been used to evaluate female breast deformities or long-term results after the removal of the pectus bar. Both questionnaires that were administered in our study, were developed by a psychologist and clinical experts with longstanding experience with chest wall deformities.
In the first questionnaire, the patients appraised their aesthetic appearance, before surgery and after pectus bar explantation, based on photographic images ([Table 1](#t1-aps-2019-00318){ref-type="table"}). The questions addressed aesthetic aspects of the breast, symmetry and strabism of the nipple-areola complex (NAC), depth of the PE deformity (funnel), and scar appearance. The same photographic images of all patients were independently evaluated by a panel of 24 persons at different levels of medical training with an equal sex distribution (three male/female medical students, three male/female residents, three male/female attending surgeons), using the same questionnaire. The evaluation by medical personnel was performed twice to assess internal consistency. The interval between the first and second evaluation was 3 months. During each session, the images were presented on a screen and the questionnaire was filled out simultaneously by all investigators.
The second patient questionnaire (questionnaire 2) was a validated standardized psychological assessment administered to patients after surgery ([Table 2](#t2-aps-2019-00318){ref-type="table"}) that addressed changes in behavior, well-being, and lifestyle.
Patients and patient images
---------------------------
Between August 2002 and February 2015, 43 female patients underwent surgical correction for PE utilizing the MOVARPE technique (n = 25) \[[@b8-aps-2019-00318]\], the MIRPE technique (n = 8) \[[@b7-aps-2019-00318]\], custom-made silicone implants (n = 6), and other procedures (n = 4). The MIRPE technique, which was first described by Nuss, involves remodeling of the anterior chest wall by implantation of a retrosternal metal bar \[[@b7-aps-2019-00318]\]. The MOVARPE procedure is a modification of the MIRPE technique with a semiopen approach with additional skin incisions in the inframammary fold, rib cartilage incisions, or partial resection and horizontal sternum osteotomy \[[@b8-aps-2019-00318]\]. The MOVARPE approach was used in adult patients or adolescents with athletic dispositions and in deformities with a deep funnel and severe sternum inclination or malrotation with asymmetry, whereas in young patients with a symmetrical depression the MIRPE technique was used. During the study period, 30 patients had their pectus bar removed (MOVARPE group, n = 23; MIRPE group, n = 7). The bar was removed after an average of 20 months (range, 12--35 months) in the MOVARPE group and 30 months (range, 14--36 months) in the MIRPE group. Patients who did not undergo bar removal were not included in the study.
All patients underwent preoperative cardiopulmonary function tests, including spirometry and echocardiography. Nine patients (50%) showed a functional deficit, which was the indication for surgery. Of the remaining nine patients, surgery was indicated in three for psychological reasons and in six for purely aesthetic reasons. In routine clinical practice, no postoperative cardiopulmonary evaluations were performed.
Of the 30 patients who were initially screened, 18 patients (60%) returned the questionnaire and were included in the study. Their mean age was 20 years (range, 14--41 years) at the time of initial surgery. The interval from pectus bar removal to the last evaluation was 28 months (range, 5--92 months). The follow-up included a clinical examination and photographic documentation in frontal, oblique, and sagittal views.
Appraisal of aesthetic appearance (questionnaire 1)
---------------------------------------------------
Questionnaire 1 was created by a group of clinical experts with long-term experience in the treatment of chest deformities. The experts designed the questionnaire based on published instruments regarding outcome evaluations after breast surgery, and specifically modified the questionnaire for PE deformities \[[@b11-aps-2019-00318],[@b12-aps-2019-00318]\]. The questionnaire included 10 items addressing the size and form of the breast, the position and symmetry of the NAC, the position and symmetry of the thoracic wall, the degree of strabism of the NAC, and the appearance and position of the scar. The questions were answered on a visual analogue scale from 1 (very poor) to 100 (very good) ([Table 1](#t1-aps-2019-00318){ref-type="table"}).
Psychological appraisal (questionnaire 2)
-----------------------------------------
We modified a German-language standardized psychological test \[[@b13-aps-2019-00318]\]. The original questionnaire consisted of 42 questions assessing the subjectively perceived status of a patient in a comparative manner. High scores signified increased relaxation, serenity, and optimism, while low scores signified increased tension, insecurity, and pessimism. A psychologist with longstanding experience with patients with chest wall deformities condensed this questionnaire to the 10 questions that were most relevant for our study.
Statistical analysis
--------------------
Data are represented by the percentage of a rating or as the median and interquartile range. In comparisons, the Wilcoxon signed-rank test for nonparametric testing of paired samples was applied. When analyzing differences between patients and professionals, the latter were grouped together with a single median score. SPSS version 19.0 (IBM Corp., Armonk, NY, USA) and MedCalc version 16.8.4 (MedCalc Software, Ostend, Belgium) were used.
RESULTS
=======
Overall impressions of preoperative aesthetic appearance
--------------------------------------------------------
Patients rated their preoperative aesthetic appearance more poorly than the medical professionals in all categories ([Table 3](#t3-aps-2019-00318){ref-type="table"}, [Fig. 1A](#f1-aps-2019-00318){ref-type="fig"}). The appearance of the funnel received the worst rating from the patients (median rating, 9), compared to a median rating of 27 by the professionals. Thorax symmetry and breast symmetry were rated as 13 and 12, respectively, by the patients, compared to ratings of 42 and 37, respectively, by the professionals.
Overall impressions of postoperative aesthetic appearance
---------------------------------------------------------
Patients rated their overall postoperative appearance as 90 on average, whereas professionals reported a rating of 72. In all aspects, patients' ratings of their own postoperative results were slightly higher than the ratings of the medical investigators, indicating a high level of satisfaction ([Table 3](#t3-aps-2019-00318){ref-type="table"}, [Fig. 1B](#f1-aps-2019-00318){ref-type="fig"}). [Fig. 2](#f2-aps-2019-00318){ref-type="fig"} shows an example of a patient who rated her preoperative appearance more poorly than the medical professionals, and was in agreement with the medical personnel in her postoperative rating.
Differences in medical professionals' evaluations between the first and second ratings
--------------------------------------------------------------------------------------
There was a strong consensus between the duplicated ratings, concerning both preoperative and postoperative ratings ([Table 4](#t4-aps-2019-00318){ref-type="table"}). In the preoperative ratings the median results for symmetry of the breast and strabism were equal between the first and second evaluations (37 vs. 37 and 41 vs. 41, respectively), while for the other questions the differences were minimal (preoperative question 1: 42 vs. 44, question 2: 46 vs. 45, question 5: 27 vs. 29). Similar results were seen in the postoperative evaluations, with only minimal differences in median ratings.
Sex differences
---------------
There were statistically significant sex differences between the groups. Overall, male evaluators assigned lower ratings than females ([Table 5](#t5-aps-2019-00318){ref-type="table"}).
Differences in medical education
--------------------------------
There were only minor differences in evaluations according to the different stages of medical education. Medical students' results showed slightly better evaluations preoperatively and slightly worse evaluations postoperatively compared to all other groups.
Questionnaire 2: Psychological questionnaire
--------------------------------------------
The psychological questionnaire demonstrated a clear benefit of surgical correction of PE in terms of psychological well-being. The design of the questionnaire allowed ratings of --3 to +3, and ratings of ≥ 1 were interpreted as indicating a positive development. Thus, 28% of patients answered 8 to 10 of the 10 questions with a positive result, 39% reported positive results on 3 to 7 of the 10 questions, and 33% of patients answered only 0 to 2 of the 10 questions positively. Regarding specific questions, 89% were satisfied with themselves after surgical correction (question 1), 83% no longer suffered from insecurity (question 4), and 78% no longer felt that they were in a battle with themselves (question 10) ([Table 2](#t2-aps-2019-00318){ref-type="table"}).
Summary of results
------------------
In summary, the patients rated their preoperative aesthetic deformities far more poorly than the medical evaluators, whereas postoperatively, patients rated their appearance equal to or slightly better than the other evaluators. In our analysis of the evaluators, we found only minor differences in the evaluations of male versus female professionals and little impact of their degree of medical education. The psychological questionnaire demonstrated a clear improvement in well-being as a result of surgery.
DISCUSSION
==========
In the present study, we focused on the aesthetic appearance, quality of life, and body image of female adult and adolescent patients before and after PE repair. Most previous studies concentrated on this deformity in males \[[@b11-aps-2019-00318],[@b12-aps-2019-00318],[@b14-aps-2019-00318]\] or even excluded female subjects from their studies \[[@b15-aps-2019-00318]\]. We focused on a female cohort, since the secondary breast deformity in females aggravates their symptoms and has a pronounced effect on body image. Since PE has been shown to cause only minor functional deficits \[[@b16-aps-2019-00318]\], insurance coverage is not always provided and the surgical correction of PE is sometimes still dismissed as a merely aesthetic procedure. However, many studies have clearly shown that the PE deformity has a significant influence on psychosocial development and that the benefits of surgical correction are not limited to aesthetic improvements \[[@b11-aps-2019-00318],[@b12-aps-2019-00318],[@b14-aps-2019-00318],[@b17-aps-2019-00318]\].
Our study demonstrated that female patients rated their preoperative deformities far worse than medical evaluators. Postoperatively, patients rated their appearance equal or slightly better than the medical evaluators. This indicates that the PE deformity strongly impacted patients' self-perception and that surgical correction successfully led to a self-evaluation that was in accordance with the other evaluators. We found only minor differences in the evaluations of male or female professionals and little impact of their degree of medical education.
Many reports about psychological health in PE patients have examined largely pediatric or adolescent patient series and have mainly included males \[[@b4-aps-2019-00318],[@b11-aps-2019-00318],[@b12-aps-2019-00318],[@b17-aps-2019-00318],[@b18-aps-2019-00318]\]. The present study focused on both adolescent and adult female patients, ranging from 14 to 41 years of age. Studies have shown significant improvements in psychological and psychosocial well-being after PE repair in children \[[@b5-aps-2019-00318],[@b12-aps-2019-00318]\] and young adults \[[@b11-aps-2019-00318]\], but the results of these studies were not based exclusively on females. Our psychological questionnaire confirmed the improvements in quality of life and psychological well-being after surgery, specifically in female patients. As stated previously, this is of particular importance because the PE deformity leads to an overall more severe aesthetic distortion in female patients than in male patients, as a result of the resultant breast deformity. Furthermore, previous investigations have usually evaluated patients before removal of the pectus bar \[[@b5-aps-2019-00318],[@b11-aps-2019-00318],[@b12-aps-2019-00318]\]. Several investigators have pointed out that data on quality of life after bar removal were not available, although these data are needed to justify performing the Nuss or MOVARPE procedure for aesthetic reasons. So far, only three reports have included postoperative evaluations, and all of these demonstrated high patient satisfaction with the results of the Nuss procedure after removal of the pectus bar \[[@b18-aps-2019-00318]-[@b20-aps-2019-00318]\]. To our knowledge, our study is the first to evaluate female PE patients' quality of life and satisfaction after bar explantation.
Many studies on aesthetic outcomes after breast surgery have confirmed that patients and medical professionals have different perceptions of the results \[[@b10-aps-2019-00318],[@b21-aps-2019-00318]\]. Previously, we conducted a study comparing the aesthetic evaluations of patients, medical professionals, and laypersons after implant-based breast reconstruction \[[@b9-aps-2019-00318]\]. In oncologic patients, we showed that patients rated their postoperative results significantly more positively than the other evaluators. Similar findings have also been reported in other studies on breast cancer patients, indicating that the differences in evaluations among patients, physicians, and laypeople are not based only on aesthetic criteria; instead, they are multifactorial \[[@b10-aps-2019-00318]\]. These results in oncologic patients are in contrast to the results of the current study on PE repair, where only the preoperative evaluations showed different ratings between patients and the other evaluators. The postoperative evaluations yielded similar aesthetic ratings in all groups. In our view, this may confirm that PE repair corrects not only an aesthetic deformity, but also has a positive influence on patients' self-perception.
Interestingly, a study by Kim et al. \[[@b19-aps-2019-00318]\], evaluating satisfaction after pectus bar removal in 39 children through a questionnaire, came to a different conclusion. The operative results were evaluated by the patients and their parents, as well as by two surgeons and two nurses. The surgical team's scores for overall satisfaction were higher than the patients' and/or parents' scores, and the same pattern was found for the scores for recommendation of the surgery to others. The authors suggested that preoperative expectations different among patients, parents, and the surgical team, and that a sufficient clarification should therefore be provided prior to surgery. We agree that patients and their parents should be thoroughly informed before surgery in order to ensure that they have realistic expectations regarding outcomes.
Patient satisfaction has become an important tool as a measure of the quality and outcomes of medical care \[[@b22-aps-2019-00318]\]. A review of patients' evaluations of medical services that included results from over 50 studies showed that the average percentage of satisfied patients was about 78% \[[@b23-aps-2019-00318]\]. The most common assumption is that satisfaction is defined as an "evaluation based on the fulfillment of expectations" \[[@b24-aps-2019-00318]\]. Various studies have tested this theory and have come to the conclusion that there is evidence to suggest that patients' expectations are involved in evaluations, but not in a simplistic way \[[@b22-aps-2019-00318],[@b25-aps-2019-00318]\]. Expectations are modified throughout the process of care and influence evaluations. A discrepancy model has been proposed, suggesting that dissatisfaction is only expressed when an extreme negative event occurs, and that the expression of dissatisfaction is likely an indicator that something went wrong \[[@b24-aps-2019-00318]\]. Therefore the high satisfaction rates in most studies are a questionable measure of medical quality, and the satisfaction questionnaires used in studies are still based on an insufficient knowledge of how and what patients evaluate \[[@b24-aps-2019-00318]\].
In accordance with these points, objective measures were analyzed in our study, and differences were still present between medical evaluators and patients. This suggests that factors beyond objective criteria influenced the evaluations.
Since self-perception is a major contributor to therapeutic decision-making, a systematic evaluation of body image should be included in the assessment of patients with thoracic wall deformities. A multidisciplinary approach, including a psychological evaluation, is mandatory for the treatment of PE deformities, and psychological screening should routinely be carried out in all patients with thoracic wall deformities. The careful selection of patients following multidisciplinary preoperative management is important for preventing disappointing results.
The limitations of this study are its retrospective data acquisition nature and the small number of patients included, which limited our ability to draw generalizable conclusions. Nonetheless, this study presents the largest experience to date, with the longest follow-up of female patients who underwent PE repair with retrosternal pectus bar implantation. Further work is needed to verify that the information the current study obtained from a small number of patients is representative of a larger population.
In summary, our results show that female patients with a PE deformity and secondary breast deformity can greatly benefit from surgical correction, with high levels of satisfaction and improvements in their quality of life. We therefore conclude that corrective surgery of PE has a positive effect on patients' physical and psychological well-being. Psychological improvement through aesthetic correction of the deformity is an indication for surgery.
**Conflict of interest**
No potential conflict of interest relevant to this article was reported.
**Ethical approval**
The study was approved by the Institutional Review Board of the Medical University of Innsbruck (IRB No. UN4960) and performed in accordance with the principles of the Declaration of Helsinki. Written informed consents were obtained.
**Patient consent**
The patients provided written informed consent for the publication and the use of their images.
**Author contribution**
Conceptualization: Wolfram D, C Mayerl C, Schwabegger AH. Data curation: Morandi EM, Verstappen R, Edlinger M, Celep E. Formal analysis: Edlinger M, Wachter T. Methodology: Wolfram D, Kinzl J, Del Frari B. Project administration: Wolfram D, Schwabegger AH, Djedovic G. Visualization: Morandi EM, Del Frari B. Writing original draft: Wachter T, Morandi EM, Del Frari B. Writing - review & editing: Wolfram D, Morandi E.
{#f1-aps-2019-00318}
{#f2-aps-2019-00318}
######
Questionnaire 1 (for aesthetic evaluation), preoperative and postoperative
----------------------------- ----- ----------------------------------------------------------------------------------------------------------------
Preoperative questionnaire
Q1: How do you rate the symmetry of the chest before the surgery?
Q2: How do you rate the symmetry of the nipples/areolas before the surgery?
Q3: How do you rate the symmetry of the breasts before the surgery?
Q4: How do you rate the extent of strabism (looking towards each other) of the nipples/areolas before the surgery?
Q5: How do you rate the extent of the funnel of the chest before the surgery?
Postoperative questionnaire
Q1: How do you rate the symmetry of the chest after the surgery?
Q2: How do you rate the symmetry of the nipples/areolas after the surgery?
Q3: How do you rate the symmetry of the breasts after the surgery?
Q4: How do you rate the extent of strabism (looking towards each other) of the nipples/areolas after the surgery?
Q5: How do you rate the position of the scars?
Q6: How do you rate the appearance of the scars?
Q7: How do you rate the extent of the funnel of the chest after the surgery?
Q8: How do you rate the appearance of the chest after the surgery?
----------------------------- ----- ----------------------------------------------------------------------------------------------------------------
Questionnaire for the aesthetic evaluation: appraisal of various aspects of pectus excavatum deformity and repair, based on standardized preoperative photographs, rated on a visual analogue scale (1=very poor; 100=very good).
######
Questionnaire 2 and results of the psychological evaluation
Item Positive rate (%)
------ ----------------------------------------------------------------- ----
Q1: I am satisfied with myself and have fewer worries about myself. 89
Q2: I am calmer inside. 67
Q3: I am more cheerful. 72
Q4: I do not suffer from insecurity anymore. 83
Q5: I am more confident in my interactions with other people. 56
Q6: I feel freer. 89
Q7: I am calmer and more balanced. 78
Q8: I now better know what is important to me. 61
Q9: I feel less insecure in communication with others. 67
Q10: I am no longer in a battle with myself. 78
Patients' appraisal of various aspects of their psychological postoperative health on a 7-point scale from --3 (very negative) to +3 (very positive) with the percentage (of n=18) of positive ratings (rating ≥1).
######
Results of questionnaire 1: preoperative and postoperative evaluations by patients and medical professionals
Patients, median (IQR) Professionals, median (IQR) P-value
-------------------------- ------------------------ ----------------------------- ---------
Preoperative evaluation
Q1: Thorax symmetry 13 (5--21) 42 (24--57) 0.001
Q2: NAC symmetry 20 (8--40) 46 (31--64) 0.012
Q3: Breast symmetry 12 (4--28) 37 (22--63) 0.001
Q4: NAC strabism 16 (10--47) 41 (29--54) 0.098
Q5: Funnel appearance 9 (3--29) 27 (17--34) 0.067
Postoperative evaluation
Q1: Thorax symmetry 76 (69--91) 72 (61--77) 0.038
Q2: NAC symmetry 73 (61--90) 68 (58--78) 0.276
Q3: Breast symmetry 76 (62--89) 62 (52--78) 0.071
Q4: NAC strabism 81 (70--93) 75 (68--82) 0.266
Q5: Scar position 92 (78--97) 82 (75--86) 0.047
Q6: Scar appearance 87 (71--96) 79 (74--84) 0.085
Q7: Funnel appearance 90 (71--94) 77 (71--83) 0.052
Q8: Overall appearance 90 (66--95) 72 (66--79) 0.112
Aesthetic appraisal of various aspects of pectus excavatum deformity and repair by patients (n=18) and medical professionals (n=24), rated on a visual analogue scale (1=very poor; 100=very good).
IQR, interquartile range; NAC, nipple-areolar complex.
######
Internal consistency based on two timely separated evaluations
Professionals P-value
-------------------------- --------------- ------------- -------
Preoperative evaluation
Q1: Thorax symmetry 42 (24--57) 44 (32--54) 0.276
Q2: NAC symmetry 46 (31--64) 45 (35--65) 0.297
Q3: Breast symmetry 37 (22--63) 37 (28--62) 0.098
Q4: NAC strabism 41 (29--54) 41 (30--52) 0.266
Q5: Funnel appearance 27 (17--34) 29 (18--37) 0.383
Postoperative evaluation
Q1: Thorax symmetry 72 (61--77) 74 (67--77) 0.037
Q2: NAC symmetry 68 (58--78) 69 (63--76) 0.050
Q3: Breast symmetry 62 (52--78) 69 (61--78) 0.002
Q4: NAC strabism 75 (68--82) 74 (68--80) 0.636
Q5: Scar position 82 (75--86) 80 (67--85) 0.485
Q6: Scar appearance 79 (74--84) 77 (63--84) 0.407
Q7: Funnel appearance 77 (71--83) 78 (69--82) 0.924
Q8: Overall appearance 72 (66--79) 75 (68--80) 0.163
Replication of aesthetic appraisal of various aspects of pectus excavatum deformity and repair by medical professionals, rated on a visual analogue scale (1=very poor; 100=very good).
IQR, interquartile range; NAC, nipple-areolar complex.
######
Sex differences in evaluations
Professionals P-value
-------------------------- --------------- ------------- ----------
Preoperative evaluation
Q1: Thorax symmetry 38 (23--51) 45 (34--59) 0.010
Q2: NAC symmetry 50 (34--65) 49 (30--64) 0.744
Q3: Breast symmetry 35 (24--61) 36 (21--60) 0.777
Q4: NAC strabism 42 (29--56) 34 (23--53) 0.013
Q5: Funnel appearance 26 (14--34) 29 (19--35) 0.124
Postoperative evaluation
Q1: Thorax symmetry 75 (66--79) 69 (64--74) 0.019
Q2: NAC symmetry 71 (57--80) 66 (56--76) 0.109
Q3: Breast symmetry 64 (54--80) 59 (48--70) 0.004
Q4: NAC strabism 80 (75--84) 69 (65--79) \< 0.001
Q5: Scar position 84 (76--86) 79 (73--83) 0.122
Q6: Scar appearance 81 (74--87) 75 (69--80) 0.010
Q7: Funnel appearance 79 (72--84) 73 (66--79) 0.006
Q8: Overall appearance 75 (70--80) 70 (64--77) \< 0.001
Sex differences in the aesthetic appraisal of various aspects of pectus excavatum deformity and repair by medical professionals compared to patients' ratings, rated on a visual analogue scale (1=very poor; 100=very good).
IQR, interquartile range; NAC, nipple-areolar complex.
[^1]: The authors thank Philip Lichtenstein, MD, from the Department of Radiology, Columbia University (New York, NY, USA), a native English speaker, for assistance with editing the English language of this paper.
|
{
"pile_set_name": "PubMed Central"
}
|
Yann de Fabrique
Yann de Fabrique (his full name is Yann de Fabrique Saint-Tours), born on 7 July 1973, is a French former swimmer who competed in the 1992 Summer Olympics and in the 1996 Summer Olympics.
His parents were both born in Martinique (FWI).
References
Category:1973 births
Category:Living people
Category:French male swimmers
Category:Male freestyle swimmers
Category:Olympic swimmers of France
Category:Swimmers at the 1992 Summer Olympics
Category:Swimmers at the 1996 Summer Olympics
Category:European Aquatics Championships medalists in swimming
Category:Universiade medalists in swimming
Category:Universiade gold medalists for France
Category:Universiade silver medalists for France
|
{
"pile_set_name": "Wikipedia (en)"
}
|
CREATE OR REPLACE VIEW PoolableElementView
AS SELECT
d.aliquotId,
d.name AS aliquotName,
d.alias AS aliquotAlias,
l.dnaSize AS aliquotDnaSize,
d.concentration AS aliquotConcentration,
d.concentrationUnits AS aliquotConcentrationUnits,
d.volume AS aliquotVolume,
d.volumeUnits AS aliquotVolumeUnits,
d.ngUsed AS aliquotNgUsed,
d.volumeUsed AS aliquotVolumeUsed,
d.identificationBarcode AS aliquotBarcode,
d.lastUpdated AS lastModified,
d.creationDate AS created,
d.creator AS creator,
d.targetedSequencingId AS targetedSequencingId,
d.lastModifier AS lastModifier,
d.preMigrationId AS preMigrationId,
d.discarded AS discarded,
d.detailedQcStatusId AS detailedQcStatusId,
d.detailedQcStatusNote AS detailedQcStatusNote,
dist.distributed AS distributed,
l.libraryId AS libraryId,
l.name AS libraryName,
l.alias AS libraryAlias,
l.description AS libraryDescription,
l.identificationBarcode AS libraryBarcode,
l.lowQuality AS libraryLowQuality,
l.platformType AS platformType,
l.paired AS libraryPaired,
sel.name AS librarySelectionType,
strat.name AS libraryStrategyType,
s.sampleId AS sampleId,
s.name AS sampleName,
s.alias AS sampleAlias,
s.description AS sampleDescription,
s.accession AS sampleAccession,
s.sampleType AS sampleType,
s.sequencingControlTypeId AS sampleSequencingControlTypeId,
p.projectId AS projectId,
p.name AS projectName,
p.shortName AS projectShortName,
p.alias AS projectAlias,
sub.subprojectId AS subprojectId,
sub.alias AS subprojectAlias,
sub.priority AS subprojectPriority,
box.alias AS boxAlias,
box.name AS boxName,
box.identificationBarcode AS boxIdentificationBarcode,
box.locationBarcode AS boxLocationBarcode,
dbp.position AS boxPosition,
d.libraryDesignCodeId AS libraryDesignCodeId
FROM LibraryAliquot d
JOIN Library l ON l.libraryId = d.libraryId
JOIN Sample s ON s.sampleId = l.sample_sampleId
JOIN Project p ON p.projectId = s.project_projectId
LEFT JOIN Subproject sub ON sub.subprojectId = s.subprojectId
LEFT JOIN LibrarySelectionType sel ON sel.librarySelectionTypeId = l.librarySelectionType
LEFT JOIN LibraryStrategyType strat ON strat.libraryStrategyTypeId = l.libraryStrategyType
LEFT JOIN LibraryAliquotBoxPosition dbp ON dbp.aliquotId = d.aliquotId
LEFT JOIN Box box ON box.boxId = dbp.boxId
-- Note: LibraryAliquotDistributionView is created in 05_view_box.sql
JOIN LibraryAliquotDistributionView dist ON dist.aliquotId = d.aliquotId;
CREATE OR REPLACE VIEW ListPoolView AS
SELECT
p.poolId,
p.name,
p.alias,
p.identificationBarcode,
p.description,
p.platformType,
p.creator,
p.created,
p.creationDate,
p.lastModifier,
p.lastModified,
p.concentration,
p.concentrationUnits,
p.discarded,
dist.distributed,
b.boxId,
b.name AS boxName,
b.alias AS boxAlias,
b.locationBarcode AS boxLocationBarcode,
bp.position AS boxPosition
FROM Pool p
LEFT JOIN BoxPosition bp ON bp.targetType = 'POOL' AND bp.targetId = p.poolId
LEFT JOIN Box b ON b.boxId = bp.boxId
JOIN PoolDistributionView dist ON dist.poolId = p.poolId;
CREATE OR REPLACE VIEW ListPoolView_Element AS
SELECT
link.poolId,
lib.libraryId,
la.aliquotId,
la.name,
la.alias,
la.dnaSize,
lib.lowQuality,
sam.project_projectId AS projectId,
sp.alias AS subprojectAlias,
sp.priority AS subprojectPriority,
ident.consentLevel
FROM Pool_LibraryAliquot link
JOIN LibraryAliquot la ON la.aliquotId = link.aliquotId
JOIN Library lib ON lib.libraryId = la.libraryId
JOIN Sample sam ON sam.sampleId = lib.sample_sampleId
LEFT JOIN Subproject sp ON sp.subprojectId = sam.subprojectId
LEFT JOIN SampleHierarchy sh ON sh.sampleId = sam.sampleId
LEFT JOIN Sample ident ON ident.sampleId = sh.identityId;
|
{
"pile_set_name": "Github"
}
|
Q:
Difference Between and when to select which when they fuse. "меня" V.S "я"
I am not only confused with the overall definition and distinction between "меня" and "я" but also when to use which when 2 different phrases containing one of them each merge together.
As an example:
Here we have "я бы" = I would
Then we have "мне хотелось" = "I wanted"
So, when conforming the phrase "I would like" as in "restaurant ordering scenarios" what comes up is:
("мне хотелось бы" why not "я хотелось бы"?)
Also from my online russian learning courses seems that this last one (мне хотелось бы), is english translated as "I would like" but that doesn't make sense to me since "to like / to enjoy" is "нравится".
Would the correct way to translate it be "I would want" instead ?
Thanks.
A:
There's no literal English analog of мне хотелось бы, because it's an impersonal form, so literal combination of words in their cases (like this: мне = to me; бы = would; хотелось = it was a wish) would bring to a strange English phrase, that is never used in the same situation: it would be a wish to me. That is why both мне хотелось бы and я бы хотел are translated as I would like.
|
{
"pile_set_name": "StackExchange"
}
|
The present invention relates to a substrate holding apparatus in which a substrate is held when polished for flattening a surface thereof. The present invention also relates to a polishing apparatus comprising the above-mentioned substrate holding apparatus.
A semiconductor device fabricating process comprises forming a thin film layer on a wafer and forming minute patterns and holes in the layer. This process is repeated until a desired number of circuit layers are formed on the wafer. Therefore, raised and recessed portions are created on or added to the surface of the wafer after formation of each circuit layer. In recent years, semiconductor devices have become increasingly minute and element structures of semiconductor devices have become complicated. Further, there is a tendency to increase the number of circuit layers for logic type devices. As a result, raised and recessed portions on the surface of a semiconductor device increase in number and a difference in height between these portions also increases. This leads to a problem such that during formation of a film on the wafer, an extremely thin film is formed over an undulating area containing the raised and recessed portions on the wafer and breaks in a circuit and an electrical insulation defect between circuit layers are likely to occur, leading to a lowering of product quality and a lowering of yield. Although semiconductor devices can operate normally during an initial period of operation, they are not reliable when used over a long period of time.
Raised and recessed portions on the wafer are also problematic in a lithography process. That is, when an exposure surface of the wafer contains raised and recessed portions, the lenses of an exposure system partially become out of focus, so that formation of minute patterns becomes difficult.
For these reasons, the techniques for surface flattening in fabricating semiconductor devices have been increasingly becoming important. Of various surface flattening techniques, the most important technique is CMP (chemical mechanical polishing), which comprises polishing by using a polishing apparatus, in which while an abrasive liquid containing abrasive particles of silica (SiO2) or the like is supplied onto a polishing surface of a polishing pad, a semiconductor wafer is slidably engaged with the polishing surface.
Conventionally, the polishing apparatus of the above-mentioned type comprises a polishing table including a polishing pad having a polishing surface and a wafer holder for holding a semiconductor wafer. The wafer holder is adapted to hold a semiconductor wafer and press the wafer against the polishing table under a predetermined pressure. The wafer holder and the polishing table are moved relative to each other so that the semiconductor wafer is slidably engaged with the polishing surface, to thereby polish the wafer to a flat and mirror-finished surface.
In the above-mentioned polishing apparatus, when a relative pressure generated between the semiconductor wafer and the polishing surface of the polishing pad is not uniform over an entire surface of the wafer, insufficient or excessive polishing is likely to occur, depending on the pressure acting on each part of the wafer. Therefore, in order to apply a uniform pressure to an entire surface of the wafer, an elastic membrane made of rubber is provided on the wafer holder on a surface thereof for holding a wafer, and a fluid pressure such as air pressure is applied to a back surface of the elastic membrane. In this case, a circumferential edge of the wafer surface corresponds to a boundary between a contact portion and a non-contact portion of the wafer relative to the polishing surface. Since the polishing pad is elastic, the pressure applied to a portion around the circumferential edge of the wafer surface becomes non-uniform, so that only the circumferential edge of the wafer is polished in an excessive amount, and the wafer is caused to have a “dull” edge.
As a countermeasure, it has been proposed to use a wafer holder in which a guide ring or retainer ring for holding an outer circumferential edge of the wafer presses the polishing surface at a position outside the wafer. In this wafer holder, the retainer ring is pressed against the polishing surface under fluid pressure such as air pressure.
FIG. 14 is a schematic illustration of a wafer holder of the above-mentioned type, in which a fluid pressure is applied to a wafer so as to press the wafer against a polishing surface, and the fluid pressure is also applied to a retainer ring so as to press the retainer ring against the polishing surface.
As shown in FIG. 14, a wafer holder 50 comprises: a wafer holder body 51 defining an inner space; a wafer pressurizing mechanism 52 contained in the inner space of the wafer holder body 51 and adapted to press a semiconductor wafer W against a polishing surface 61 of a polishing table 60; a retainer ring 53 provided so that it is vertically movable relative to the wafer holder body 51 and adapted to hold an outer circumferential edge of the wafer W; and a retainer ring pressurizing mechanism 54 for pressing the retainer ring 53 against the polishing surface 61.
The wafer pressurizing mechanism 52, although not shown in detail, comprises an elastic membrane member which is made of an elastic material such as rubber and is connected to the wafer holder body 51. A pressurized fluid such as pressurized air is supplied to the inside of the elastic membrane member so that the wafer W is pressed against the polishing surface 61 under fluid pressure. The retainer ring pressurizing mechanism 54, although not shown in detail, also comprises an elastic membrane member which is made of an elastic material such as rubber and is connected to the wafer holder body 51. A pressurized fluid such as pressurized air is supplied to the inside of the elastic membrane member so that the retainer ring 53 is pressed against the polishing surface 61 under fluid pressure. The wafer holder body 51 is connected to a drive shaft 55 and the drive shaft 55 is adapted to be vertically moved by a lifting mechanism such as an air cylinder.
The lifting mechanism such as an air cylinder connected to the drive shaft 55 is operated so as to move the wafer holder body 51 as a whole to a position close to the polishing table 60. While the wafer W is held in proximity to the polishing surface 61, the pressurized fluid is supplied under a predetermined pressure to the wafer pressurizing mechanism 52, to thereby press the wafer W against the polishing surface 61 of the polishing table 60. The pressure applied to the wafer W during polishing is adjusted to a desired value by adjusting the pressure of the pressurized fluid supplied to the wafer pressurizing mechanism 52. On the other hand, the pressurized fluid is supplied under a predetermined pressure to the retainer ring pressurizing mechanism 54, to thereby press the retainer ring 53 against the polishing surface 61 of the polishing table 60.
Since the wafer W is pressed against the polishing surface 61 by using a fluid pressure, it is possible to obtain a uniform pressure distribution across an entire surface of the wafer W from the center to the circumferential edge thereof. This enables uniform polishing of the entire surface of the wafer W. Further, during polishing, a pressure substantially equal to that applied to the wafer W is applied to the retainer ring 53 through the retainer ring pressurizing mechanism 54, so that the polishing surface of the polishing pad outside the wafer W is pressed under a pressure substantially equal to that of the wafer W. Therefore, a uniform pressure distribution can be obtained continuously across an area from the center of the wafer W to an outer circumferential portion of the retainer ring 53 outside the wafer W. Therefore, excessive or insufficient polishing at the circumferential edge of the wafer W can be prevented.
In the above-mentioned conventional wafer holder in which both the wafer and the retainer ring are pressed under fluid pressure, the retainer ring is capable of moving in either a vertical (or perpendicular) direction or a lateral (or radial) direction relative to the wafer holder body. That is, the retainer ring is capable of moving independently of the wafer holder body. Movement of the retainer ring affects uniformity in the polishing of an outer circumferential portion of the wafer surface. Although vertical movement of the retainer ring is necessary for polishing, lateral movement of the retainer ring is unnecessary. Rather, lateral movement of the retainer ring is undesirable because it varies the distance between the retainer ring and the circumferential edge of the wafer surface and impairs uniformity and stability in the polishing of the outer circumferential portion of the wafer surface.
Further, in the conventional wafer holder, since the surface of the wafer holder for holding a wafer is covered with the elastic membrane, it is required to form, for example, a suction cup-like configuration in the elastic membrane so as to hold a wafer during transfer thereof. When a wafer is held by the elastic membrane having a suction cup-like configuration, warpage or deformation of the wafer occurs. Due to warpage of the wafer, the wafer can be broken during transfer thereof or a device structure formed on the wafer can be damaged. Further, since the wafer is held by indirect contact with the wafer holder through the elastic membrane, defects in holding of the wafer are likely to occur during transfer of the wafer, leading to a lowering of operating rate of the wafer holder and a lowering of yield of wafers.
Further, in chemical mechanical polishing (CMP) utilizing an elastic polishing pad and an abrasive liquid (slurry), the following problem arises. That is, when a wafer surface having raised and recessed portions is polished, the raised portions are polished in preference to the recessed portions during an initial period of polishing, but after the raised portions are polished by a certain amount, the recessed portions are also gradually subjected to polishing (as well as the raised portions). Therefore, the difference in height between the raised portions and the recessed portions cannot be easily reduced. That is, because polishing is conducted by using a relatively soft, elastic polishing pad and a slurry type abrasive liquid containing a large amount of free abrasive particles, chemical mechanical polishing is effected on not only the raised portions, but also the recessed portions of the wafer surface. Further, the effect of polishing varies, depending on the density of raised and recessed portions.
Therefore, an attempt has been made with respect to polishing by using a polishing surface comprising fixed abrasive particles such as cerium oxide (CeO2), which are bound by using a binder such as a phenol resin. In this polishing, the polishing surface is hard as compared to the polishing pad conventionally used in chemical mechanical polishing, so that the raised portions are polished in preference to the recessed portions and the recessed portions are unlikely to be polished. Therefore, absolute flatness of the wafer can be easily obtained.
However, a wafer holder suitable for a hard polishing surface comprising fixed abrasive particles has not been developed. Generally, a conventional wafer holder for the hard polishing surface comprises a rigid wafer holder body and an elastic backing pad provided on the rigid wafer holder body adapted to be engaged with a wafer to be held by the wafer holder. Although the elastic backing pad can absorb shocks on the wafer, it is difficult for the elastic backing pad to take care of undulations on the hard polishing surface, whereby the undulations are transferred to and affects the wafer surface to be polished.
|
{
"pile_set_name": "USPTO Backgrounds"
}
|
<?xml version="1.0" encoding="utf-8"?>
<ripple xmlns:android="http://schemas.android.com/apk/res/android"
android:color="@color/ripple_color">
<item>
<selector>
<item
android:drawable="@drawable/button_white_stroke_green_p"
android:state_pressed="true" />
<item android:drawable="@drawable/button_white_stroke_green_n" />
</selector>
</item>
</ripple>
|
{
"pile_set_name": "Github"
}
|
Thesis of the french revolution
Thesis of the french revolution - Morris Slavin, Historian of the French Revolution
3 In these times, this declaration was absolute revolutionary in no other nation had all men the same rights and were declared equal.
The moment Napoleon Bonaparte assumed power in France. They were locked out for fear of what they could do. It is a scenario with many parallels in the modern political world though thankfully without the guillotine as the denouement. 9 pages Term Papers- Poetry and Song in the French Revolution The French Revolution is perhaps one of the most confusing, illogical and fascinating period of modern European history. Law and the French Revolution The French government was a complete monarchy. Furthermore worked the American society very well due to the Declaration of Independence and became so an example for the French and other European societies.
The transformation of the Russian Revolution into its monstrous has revealed the meaning of the Great State. One of the major dilemmas was the French government inability to collect the sufficient amount of taxes from the people. In addition, there was passed a bill, which was prepared by Lafayette and Jefferson and which abolish any title or rank to make all men equal. Since France had suffered financially, most people faced starvation and economic difficulties. The distinguished French historian of medicine, Jean-Charles Sournia, has said that the term deserves special attention, but he devotes no more than half a page to it. Mackintosh continues Legislators are under no obligation to retain a constitution, because it has been found quoting Burke tolerably to answer the common purposes of Government. He placed under the microscope the lives of the ordinary individuals who inhabited what today is still known as the Marais he examined how they advanced the cause of the Revolution by their agitation in various journ es and were, in turn, affected by its course. Such a situation led to the discontent of middle, lower and working class that wanted to be heard.
3 pages Term Papers- History through out time has maintained a pattern with almost all major events the French Revolution being no different.
Although French citizens like her personality and beauty a lot, but Austria and France were enemies during that time. H-Net permits the redistribution and reprinting of this work for nonprofit, educational purposes, with full and accurate attribution to the author, web location, date of publication, originating list, and H-Net Humanities Social Sciences Online. His methodology followed traditional form, with the all available evidence carefully weighed and conclusions drawn logically from the facts. Indeed, once free travel to France was resumed in 1815, the exceptional developments in French medicine during the revolutionary and Napoleonic period came to be studied by foreign students and doctors among the crowds of visitors.
As the Enlightenment took hold and ideas about the right kind of government began to spread, tension only began to increase. The Old Regime used the feudalism system left from the Middle Ages.
All these aspects had been parodied by Moli re, who created an image of the physician perhaps most justly applied to the Paris Faculty of Medicine and widely believed by many people in the eighteenth century. On his sources, see the preface and bibliography of his Left of the Mountain The Enrag s and the French Revolution Ph. Transitions should appear at the beginning of the next body paragraph. The next year the leaders of the National Convention drafted a new plan for government. From 1789-1804 there was no increase in irrigated area in Provence and the irrigation that was already in place was barely maintained.
The National Committee was made up of French men like Robespierre and were most likely all in the same socioeconomic class.
What you need to do now is start to transform this project into the form of a research paper, with a separate introduction, body, and conclusion, and intersperse your images accordingly. However, beneath the shiny surface, a storm was brewing. As one proceeds through the chapters of Human Rights and Revolutions, the suspicion arises that the connection between the two may not only be more systemic than many contributors are suggesting, but also less paradoxical.
It began from 1789 due to many major causes that included economic and political crisis and lasted till 1794. In his conclusion, Lefevbre writes, The Revolution of 1789 consisted first of all in the fall of absolute monarchy and advent of liberty henceforth guaranteed by constitutional government nor on this score can it be doubted that it was a national revolution, since the privileged orders as well as the Third Estate demanded a constitution and a regime in which individual rights would be respected, 209. tags American Revolution, French Revolution 1069 words 3. Your thesis should be between one and three sentences long and is placed at the end of your introduction.
While you have all the components of a project, you have not transitioned them into a flowing narrative, rather this reads as a list of summaries with little work done to integrate them into a larger framework, or to apply historical examination to them.
But this is only half the story and a measure which is far from reflecting discredit on the revolution.
Beginning with a country in economic turmoil and an unjust political system, France was rapidly transformed into a nation of revolt that triggered a call for revenge upon those who had done wrong.
J F Baraillon, D fense d exercer l art de gu rir sans titre authentique, 1797, p. Conversely however, The Terror as well promoted the oppression of innocent individuals from all levels of society and was hypocritical to the initial objectives of the revolution to create a just The French Revolution was the result of Louis stubbornness. The death of his wife in 1782 added to his troubles, but he got again a seat in the Congress. The inhabitants had the same rights as a free english man that meant that they were allowed to have a which decided on political issues. These hundred years were much need after the wars of Napoleon.
Corbyn s spectacular success in becoming leader of his party on an overwhelmingly high popular mandate, and against the wishes of many of the professional politicians in his own party, has other resonances with the French Revolution. A group of commoners dared storm the prison on 14th of July in 1789. The members of the College were doing no more than expressing the widely felt sentiment that in skill and in medical effectiveness surgeons were at least the equal of physicians. Over the next few years there were many administrative changes, but on the whole successive governments continued to refer medical matters to a health committee.
tags world history, social political turmoil, France 15 Works Cited 951 words 2.
|
{
"pile_set_name": "Pile-CC"
}
|
package io.cattle.platform.api.resource;
import io.cattle.platform.util.type.InitializationTask;
import io.github.ibuildthecloud.gdapi.request.resource.impl.ResourceManagerLocatorImpl;
public class ExtensionResourceManagerLocator extends ResourceManagerLocatorImpl implements InitializationTask {
@Override
public void start() {
init();
}
}
|
{
"pile_set_name": "Github"
}
|
Introduction {#Sec1}
============
Together with the advent of graphene science, an important chapter in organic chemistry has opened exploring the synthesis of extended polycyclic aromatic hydrocarbon (PAH) systems. Large PAHs, as represented by hexa-*peri*-hexabenzocoronene (HBC, **1**) with π-conjugated structures extending over 1 nm, can be regarded as structurally defined, nanoscale cutouts of graphene, designated nanographene molecules^[@CR1]--[@CR3]^. Such systems can serve as platforms both for investigating the physical and chemical properties of graphene upon heteroatom-doping, as well as for the development of semiconducting materials (see Fig. [1](#Fig1){ref-type="fig"}). Doped nanographenes are similarly relevant for the study of aromaticity and antiaromaticity, which are fundamental concepts of organic chemistry^[@CR4],[@CR5]^. As proposed by Hückel^[@CR6]^ and later by Frost and Musulin^[@CR7]^, as well as Breslow^[@CR8]^, π-conjugated cycles with \[4n + 2\] π-electrons show aromatic character with strong stabilization while cycles with \[4n\] π-electrons are antiaromatic and destabilized. Despite a few theoretical suggestions on local antiaromatic "defects", or "dopants", in graphene^[@CR9],[@CR10]^ and rapidly increasing variety of nanographene molecules, well-defined nanographene molecules with a pronounced antiaromatic ring, i.e., a local antiaromatic element, remain challenging targets to be explored.Fig. 1Chemical structures of pristine and N-doped nanographenes. Hexa-*peri*-hexabenzocoronene (HBC) **1**, periphery-doped tetraaza-HBC **2**, azacoronene derivatives **3**, and azacorannulene **4**
Nitrogen (N)-doped nanographenes have attracted considerable attention due to promising electronic and magnetic properties^[@CR11]^ offering potential applications in metal-free catalysis^[@CR12],[@CR13]^ and sensing^[@CR14],[@CR15]^. However, the existing synthetic protocols do not allow the precise control of the doping level and heteroatom positioning. The chemical nature of nitrogen dopants (i.e., pyridinic, pyrrolic, and "graphitic" N atoms) remains largely undefined^[@CR16]^, restricting the fine-tuning of the resulting properties as well as reliable structure-property relationship elucidation. In this regard, structurally well-defined N-doped nanographene molecules offer an opportunity to accurately investigate N-doping effects. Notwithstanding, N-doping of defined nanographenes has been predominantly limited to pyridinic^[@CR17]--[@CR19]^ (e.g., compound **2** ^[@CR20]^ in Fig. [1](#Fig1){ref-type="fig"}) and pyrrolic (e.g., compounds **3** ^[@CR21]^ and **4** ^[@CR22],[@CR23]^) N atoms, which only resulted in replacing benzene rings with other N-containing 6π aromatic rings, such as pyrimidines and pyrroles.
To develop nanographenes with strong antiaromatic ring dopants, we consider the possibility of incorporating a pyrazine structure inside the aromatic core, which employs two "graphitic" N atoms in the framework, thus providing additional π-electrons when compared to a benzene ring. 1,4-Disubstituted pyrazine entails intriguing chemistry, expressing 8π antiaromatic (neutral), 7π non-aromatic (cationic), or 6π aromatic (dicationic) properties, depending on its oxidation state. Thus, a pyrazine ring embedded in a carbonaceous hexagonal structure serves as a localized antiaromatic ring dopant, influencing ring and perimeter currents^[@CR24],[@CR25]^ with the potential to give rise to new types of 2D materials with unconventional semiconducting properties.
Here, we report in situ solution synthesis and spectroscopy of the partially fused, pyrazine-incorporated hexabenzoperylene (HBP) derivative **6a** as well as on-surface synthesis of the diaza-HBC **7c** with a completely embedded pyrazine core. N-doped HBP **6a** features an 8π antiaromatic core structure, as documented by ^1^H NMR and UV--vis absorption spectroscopy together with density functional theory (DFT) calculations. Notably, the solution chemistry of polycyclic aromatic azomethine ylides (PAMYs)^[@CR26]^ can be successfully adapted on the Ag(111) surface, achieving the synthesis of **7c** through surface-assisted homocoupling of the dibenzo-9*a*-azaphenalene (DBAP) salt **5c** and concomitant cyclodehydrogenation. High-resolution scanning tunneling microscopy (STM) and frequency-modulated atomic force microscopy (FM-AFM)^[@CR27]^ unambiguously elucidate the chemical structure of diaza-HBC **7c**. Moreover, computational modeling using DFT indicates that **7c** conserves its 8π character on the Ag(111) surface. Nucleus-independent chemical shift (NICS)^[@CR28]^ calculations on **6a** and **7c** reveal high positive values at the center and 1 Å above the center of the central pyrazine rings (NICS(0) and NICS(1) values, respectively), indicative of antiaromaticity. These results demonstrate the antiaromatic nature of pyrazine-doped nanographene molecules and provide an on-surface synthesis route to N-doping nanographene materials.
Results {#Sec2}
=======
Solution synthesis and characterizations {#Sec3}
----------------------------------------
The synthesis of pyrazine-embedded HBP **6a** (Fig. [2](#Fig2){ref-type="fig"}) was carried out based on the chemistry of DBAP, which has been employed to construct N-doped PAHs such as dibenzo\[*d*,*k*\]ullazines and 6*b* ^2^-azapentabenzo\[*bc*,*ef*,*hi*,*kl*,*no*\]corannulenes^[@CR22],[@CR26],[@CR29],[@CR30]^. DBAP salts **5a-c** with different substituents were first synthesized according to our previous method^[@CR26]^. They were then dimerized to form precursors **8a--c** (Fig. [2](#Fig2){ref-type="fig"}) by treatment with a large excess of tributylamine in dry dimethyl sulfoxide (DMSO) under argon at 190 °C. Oxidation of **8a** with excess 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) in dry C~2~D~2~Cl~4~ afforded diaza-HBP **6a**, as characterized by ^1^H NMR and mass spectroscopy (MS) analyses (Fig. [3](#Fig3){ref-type="fig"}). Diaza-HBP **6a** turned out to be extremely unstable probably because of the antiaromatic pyrazine core, and thus the oxidation step was performed in situ in a sealed NMR tube under exclusion of air. Oxidation of other precursors **8b** and **8c** were also attempted with the same method, but failed to provide corresponding diaza-HBP derivatives according to ^1^H NMR analysis, suggesting that the electron-donating methoxy groups of **8a** played an important role in the reaction^[@CR31]^. As shown in Fig. [3a](#Fig3){ref-type="fig"}, the disappearance of the characteristic signal from H~F~ (5.13 ppm) of **8a** indicated complete conversion of the starting material. All the signals in the aromatic region of **6a** were down-field shifted compared to those of **8a**, in agreement with the extension of the π-conjugation upon dehydrogenation. Notably, proton H~E~ of **8a** is above the neighboring benzene plane according to the optimized geometry by DFT (Fig. [4](#Fig4){ref-type="fig"}) and thus is strongly shielded. In contrast, after dehydrogenation of **8a** via DDQ, the proton H~e~ of **6a** is shifted away from the neighboring benzene into its deshielding region, in accordance with a down-field shift of H~e~ in the ^1^H NMR spectrum. Moreover, ^1^H--^1^H correlation spectroscopy (COSY) and ^1^H-^1^H nuclear Overhauser effect spectroscopy (NOESY) analyses (Fig. [3](#Fig3){ref-type="fig"}b, [c](#Fig3){ref-type="fig"}) allowed unambiguous assignments of the proton signals of **6a**. The structure was further characterized by ^1^H--^13^C heteronuclear single-quantum correlation (HSQC) and ^1^H--^13^C heteronuclear multiple-bond correlation (HMBC) spectroscopies. The chemical identity of **6a** was confirmed by high-resolution matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) MS, which showed a matching isotopic pattern starting at *m*/*z* = 590.1980, in agreement with the theoretical value of 590.1994 (Fig. [3d](#Fig3){ref-type="fig"}).Fig. 2Synthesis of diaza-HBP **6a** in solution and internally N-doped diaza-HBC **7c** upon interfacial confinement. Reagents and conditions: (i) tributylamine, DMSO, 190 °C; (ii) DDQ, C~2~D~2~Cl~4~, 100 °C; (iii) in vacuo on the Ag(111) surface, 270 °C. The calculated nucleus-independent chemical shifts (NICS) values are depicted in blue (unit: ppm), indicating the antiaromatic nature of the pyrazine ring (in parenthesis the NICS(1) value, see text and Supplementary Fig. [1](#Fig1){ref-type="fig"}) Fig. 3Structural characterizations of the in situ generated diaza-HBP **6a**. **a** ^1^H NMR spectra of **8a** and **6a** together with their structural assignments (\*C~2~HDCl~4~). Compared to precursor **8a**, the signals of **6a** were down-field shifted, corroborating the extension of the π-conjugation. **b** Aromatic region of the ^1^H--^1^H COSY spectrum of **6a** in C~2~D~2~Cl~4~. **c** Aromatic region of the ^1^H--^1^H NOESY spectrum of **6a** in C~2~D~2~Cl~4~. **d** High-resolution MALDI-TOF MS spectrum of **6a**, with an inset showing its experimental (black line) and simulated isotopic distributions (red line) Fig. 4Photophysical investigations. **a** Experimental and **b** TDDFT-simulated absorption spectra of **6a** and its precursor **8a**. The simulated spectra of **6a** showed the appearance of a new band at 577 nm, in accordance with the experimental observation. **c** X-ray single-crystal structure of **8c**. **d** DFT-optimized structure of **8a**. **e** DFT-optimized structure of **6a**
To investigate the photophysical properties of **6a**, we prepared diaza-HBP **6a** in situ by carrying out the oxidation reaction in a sealed quartz cell and recorded the absorption and photoluminescence spectra. As shown in Fig. [4a](#Fig4){ref-type="fig"}, dimer precursor **8a** exhibited an absorption maximum at *λ* ~max~ = 390 nm, and an absorption onset at ≈500 nm. After dehydrogenation, the absorption maximum barely changed, but a new band emerged in the range of 500 to 630 nm, suggesting an extended π-conjugation. To further understand the photophysical properties, DFT calculations were performed on both **8a** and **6a**. In order to determine the exact molecular geometry of **8a**, growth of single crystals of **8a--c** was pursued and a single crystal of **8c** suitable for X-ray analysis was successfully obtained, revealing an *anti*-folded structure (Fig. [4c](#Fig4){ref-type="fig"}). Thus, the *anti*-folded geometry was employed as input geometry for the DFT optimization of similar precursors (Fig. [4d](#Fig4){ref-type="fig"}). DFT studies on dehydrogenated **6a** indicate that the *anti*-folded conformation (Fig. [4e](#Fig4){ref-type="fig"}) is energetically preferred compared to the twisted one. Time-dependent DFT (TDDFT) calculations of **6a** and **8a** indicated that the emerging absorption band of **6a** can be attributed to the HOMO→LUMO + 2 transition (Fig. [4b](#Fig4){ref-type="fig"} and Supplementary Tables [1](#MOESM1){ref-type="media"}--[3](#MOESM1){ref-type="media"}). The HOMO→LUMO transition of **6a** is symmetry forbidden, and **6a** is non-emissive, in agreement with previous reports on antiaromatic systems^[@CR32],[@CR33]^.
On-surface synthesis of 7c {#Sec4}
--------------------------
Despite generation of diaza-HBP **6a** through solution synthesis, further characterization of **6a** was restricted by its low stability. In addition, attempts to oxidize intermediate product **8a--c** directly to diaza-HBC derivatives resulted in insoluble solids, which could not be characterized. This led us to adapt the solution chemistry of PAMYs to on-surface synthesis^[@CR34]--[@CR37]^, which offers today a new approach to organic chemistry. We thus deposited **5c** on Ag(111) by molecular beam evaporation (Methods section and Supplementary Fig. [2](#MOESM1){ref-type="media"}). Upon annealing to 270 °C, small quantities of diaza-HBC **7c** were detected as displayed in Fig. [5a](#Fig5){ref-type="fig"}. Unlike solution synthesis (Fig. [2](#Fig2){ref-type="fig"}), no intermediate product **8c** was identified after annealing the deposited **5c** in several ultra-high vacuum (UHV) preparations on Ag(111), Ag(100), Cu(111), and boron nitride (BN) on Cu(111), suggesting that homocoupling and dehydrogenation occur synergistically at the employed reaction temperatures. Notably, **7c** was detected on BN/Cu(111) templates^[@CR38]^ at similar annealing temperatures in trace-quantities (Supplementary Fig. [3](#MOESM1){ref-type="media"}), indicating that the on-surface coupling strategy proceeds in a metal-free fashion, further motivating studies on reaction mechanisms and on-surface synthesis on insulators.Fig. 5On-surface homocoupling of precursor **5c** yields fully planar and neutral diaza-HBC **7c** on Ag(111). **a** Scanning tunneling microscopy (STM) overview of reaction product and unreacted species on Ag(111). *V* ~s~ = 30 mV, *I* ~t~ = 10 pA. **b** Constant-height STM data of diaza-HBC with partially superposed molecular model. *V* ~s~ = 5 mV. **c** Frequency-modulated atomic force microscopy (FM-AFM) and **d** Laplace-filtered FM-AFM data reveal details in the molecular structure. Apparent C = C distance 1.0 Å (black), and N--C 1.8 Å (blue). **e** FM-AFM simulations of **7c** employing **a** neutral tip. Apparent C = C distance 1.5 Å (black), N--C 1.5 Å (blue). **f** FM-AFM simulations employing a negatively charged probe tip. Apparent C = C distance 1.3 Å (black), N-C 1.9 Å (blue). For simulation parameters see Methods. **g**--**n** Constant-height STM data (**g**, **i**, **k**, **m**) at different biases and corresponding (**h**, **j**, **l**, **n**) DFT simulations. White arrows point to N atoms oriented along the diaza-HBC symmetry axis. Scale bars (**a**) 20 Å, (**b**--**f**) 2 Å, (**g**--**n**) 5 Å
High-resolution, constant height STM (Fig. [5b](#Fig5){ref-type="fig"}) and FM-AFM (Fig. [5c, d](#Fig5){ref-type="fig"}), using CO-functionalized tips, confirmed **7c** on Ag(111). In FM-AFM data on Ag(111), the central pyrazine ring of **7c** (Fig. [5d](#Fig5){ref-type="fig"}) appears distorted. Observation of a *D* ~2*h*~ symmetry points toward the nitrogen atoms sitting at the positions indicated by white arrows. Note that the predicted DFT C=C bond length of the central pyrazine ring is 1.40 Å vs. 1.42 Å for the N--C bond, and the molecule remains planar (z ± 0.05 Å) for calculated adsorption sites on Ag(111) (Supplementary Fig. [4](#MOESM1){ref-type="media"}). Bond lengths in AFM data typically undergo distorsions^[@CR27],[@CR39]--[@CR41]^ especially under the influence of electrostatic interactions with the CO tip^[@CR42]--[@CR46]^. This causes a distorted appearance of the central pyrazine ring. Quantitative evidence of these combined effects is summarized in the AFM frequency shift simulation including a DFT-derived electrostatic potential (Fig. [5e, f,](#Fig5){ref-type="fig"} see Methods section). Note how the apparent C=C bond length in the simulated image (1.3 Å) appears shorter than the apparent N--C bond length (1.9 Å, Fig. [5f](#Fig5){ref-type="fig"}), in good agreement with the experimental data (1.0 and 1.8 Å, respectively). When a neutral tip is employed for the AFM simulation the central pyrazine ring is fully symmetric, with apparent N--C and C=C lengths of 1.5 Å (Fig. [5e](#Fig5){ref-type="fig"}), confirming that the distortions in the FM-AFM data originate mostly from electrostatic effects.
Combined scanning probe data and theoretical investigations indicate that **7c** remains neutral (conserving its pyrazine 8π state) on the surface, which is a requirement for antiaromaticity. Bader charge analysis was employed to assess molecule-substrate charge transfer. As shown by NICS calculations on isolated molecules, a neutral species of **7c** implies an 8π pyrazine unit, which is antiaromatic in the isolated state. Upon oxidation, a positively charged "7π" radical cation state of the diaza-HBC **7c** displays significantly reduced antiaromaticity and a "6π" dication species exhibits an aromatic character (Supplementary Fig. [1](#MOESM1){ref-type="media"}). On Ag(111), theoretical charge analysis finds only marginal electron transfer to the substrate (the molecule is charged positively by less than + 0.14*e* for all calculated molecular configurations on Ag(111), Supplementary Fig. [4](#MOESM1){ref-type="media"}). Moreover, when trying to enforce molecular charging by removing electrons from the calculated system, the charge at the molecule was only increased to about + 0.8*e*, when the whole 29 Å × 25 Å × 22 Å unit cell was charged by + 10*e*. These theoretical analyses indicate that **7c** remains in the neutral "8π" state on Ag(111). The good correlation of the STM experimental data with Tersoff-Hamann simulations supports the validity of the theoretical conclusions (Fig. [5g, n](#Fig5){ref-type="fig"} and see Supplementary Fig. [5](#MOESM1){ref-type="media"} for extended comparisons). It should be noted how at negative energies (voltages), the symmetry of the tunneling image follows the diaza-HBC **7c** hexagonal frame (marked by a white arrow, Fig. [5g, h](#Fig5){ref-type="fig"}). At high positive biases (between 0.9 and 2.1 eV, see Supplementary Fig. [5](#MOESM1){ref-type="media"}) the symmetry of the tunneling data is rotated (blue arrows, Fig. [5k, l](#Fig5){ref-type="fig"}) with respect to the diaza-HBC hexagonal frame, that is, tunneling is enhanced between the bay regions in the periphery of diaza-HBC. Notably, between −0.5 and 0.5 eV (Fig. [5i, j, m, n](#Fig5){ref-type="fig"}), both experiment and simulation follow the hexagonal symmetry of **7c** on Ag(111).
Having identified a neutral state of **7c** on Ag(111), comparison between **7c** on Ag(111) and the isolated antiaromatic species provides evidence on the preserved antiaromatic character of **7c** on Ag(111). Figure [6](#Fig6){ref-type="fig"} depicts the site-projected density of states (PDOS) as well as plots for the individual Kohn-Sham orbitals. Only marginal hybridization resulting in mid-gap states is observed when the isolated molecule interacts with Ag(111) (Fig. [6a](#Fig6){ref-type="fig"}, red line). The mid-gap states do not appear to significantly influence the antiaromatic character of the adsorbed molecule: the HOMO (H) and HOMO-1 (H-1) signatures of isolated **7c** are reflected undisturbed at the surface (compare Fig. [4](#Fig4){ref-type="fig"}b, c, e, f). Instead, the electronic density difference plot (Fig. [6d](#Fig6){ref-type="fig"}) shows that the mid-gap states originate from homogeneous orbital contributions, including the role of rings which are considered to be aromatic (Supplementary Fig. [1](#MOESM1){ref-type="media"}). Note that scanning tunneling spectroscopy follows the major theoretical LDOS features including a pronounced broadening (Supplementary Fig. [6](#MOESM1){ref-type="media"}).Fig. 6DFT electronic structure comparison of the isolated and Ag(111)-supported diaza-HBC **7c**. **a** Projected *p* ~z~ contributions to the density of states of isolated **7c** (blue) and on Ag(111). The Fermi energy (*E* ~F~) of the isolated molecule has been arbitrarily defined for comparison. **b**, **c** Contribution of the highest occupied molecular orbital **H** ~**ad**~ (at −0.32 eV in **a**) and (**H-1**)~**ad**~ to **7c** on Ag(111). **d** Electron density difference between **7c** on Ag(111) and the isolated molecule plus isolated substrate. Red regions represent electron density depletion, whilst blue electron density gain. Atoms are rendered translucent. **e**, **f** Highest occupied molecular orbital **H** and **H-1** of isolated **7c**. These results suggest that antiaromatic character is preserved on Ag(111)
Discussion {#Sec5}
==========
In summary, we have demonstrated the chemical synthesis of a partially fused PAH in solution and the on-surface synthesis of a fully planarized nanographene molecule with antiaromatic pyrazine dopants. Diaza-HBP **6a** has been achieved in solution and investigated by NMR, high-resolution MALDI-TOF MS and UV-vis absorption spectroscopy, while the fully cyclized diaza-HBC **7c** has been successfully fabricated on Ag(111) and characterized by STM and FM-AFM. We suggest that pyrazine rings are potential candidates to be included in 2D materials to spawn novel functional properties induced by breaking the conjugated lattice via antiaromatic ring dopants. Importantly, the chemistry of on-surface azomethine ylide homocoupling reported here opens up a new route to novel nanographene materials, especially graphene nanoribbons, on a large variety of substrates.
Methods {#Sec6}
=======
Procedure for the in situ preparation of **6a** {#Sec7}
-----------------------------------------------
Inside an argon-filled glovebox, precursor **8a** (2.0 mg, 1.0 eq.), DDQ (15 mg, 20 eq.) and C~2~D~2~Cl~4~ (0.6 mL) were added sequentially into a sealable NMR tube. The mixture was heated at 100 °C for 1 min and cooled down to room temperature, and then analyzed by NMR spectroscopy. Synthetic procedures and characterization data of **8a--c** and **6a** are provided in the [**Supplementary Methods**](#MOESM1){ref-type="media"} and Supplementary Figs. [7](#MOESM1){ref-type="media"}--[45](#MOESM1){ref-type="media"}.
STM and FM-AFM measurements {#Sec8}
---------------------------
Precursor **5c** was sublimated between 200 °C and 220 °C at 10^−9^ mbar. Data was acquired in situ using a CreaTec STM/AFM (5.5 K) under ultra-high vacuum (10^−10^ mbar). AFM data were taken in constant height mode at 0.0 V bias, using a qPlus sensor (resonance frequency ≈22 kHz, oscillation amplitude: 70 pm to 100 pm). The tungsten tip was prepared by focused-ion beam processing and in situ tip forming. For subsequent tip functionalization, carbon monoxide (CO) was dosed on the cold sample and transferred to the tip. The images were processed using the Gwyddion software package^[@CR47]^.
Computational details {#Sec9}
---------------------
DFT Calculations in Fig. [4](#Fig4){ref-type="fig"} were performed using the Gaussian 09 software package. The geometries were optimized at the B3LYP/6-311 G(d,p) level. The excitation energies were calculated by the TDDFT method at the B3LYP/6-311 G(d,p) level. NICS were calculated using the gauge invariant atomic orbital (GIAO) approach at the GIAO-B3LYP/6-311 + G(2d,p) level. NICS(1) values were averaged by two positions (above and below the plane). For Figs. [5](#Fig5){ref-type="fig"} and [6](#Fig6){ref-type="fig"}, periodic DFT calculations were performed with the VASP code^[@CR48]^, projector augmented wave potentials^[@CR49]^ and plane waves expanded to a kinetic energy cutoff of 400 eV. Exchange-correlation effects were treated by the van der Waals density functional^[@CR50]^ in the version denoted as rev-vdWDF2^[@CR51]^. The Ag(111) surface was represented by four layered slab separated by approx. 15 Å of vacuum region. Furthermore, a *p*(10×10) super cell was used together with a 2×2 *k*-point sampling. All structures were geometrically optimized until the residual forces on all atoms were smaller than 0.01 eV/Å, except for the bottom two layers of the slab, which were frozen in the bulk distance. An exhaustive conformational search identified the preferred adsorption site (Site b, Supplementary Fig. [4](#MOESM1){ref-type="media"}). The site's adsorption height was further optimized for a gas-phase molecule and found to be 3.22 Å between the molecular and metallic planes (Site c, Supplementary Fig. [4](#MOESM1){ref-type="media"}). This site was employed to simulate the STM and AFM images. STM images were simulated with the Tersoff-Hamann approximation^[@CR52]^ using the implementation by Lorente and Persson^[@CR53]^. AFM simulations were performed with a modification of the particle probe model code^[@CR42]^ by Hapala et al. The DFT-derived Hartree potential was employed to prepare an electrostatic grid which together with the molecular Lennard-Jones potential acted as the force-field for simulating the frequency shift of a probe C = O particle as detailed in ref. ^[@CR46]^. The simulations involved oscillating a probe particle with an amplitude of 5 Å at a height of 7.7 Å above the substrate. The terminal oxygen was given a negative charge of - 0.005*e*. The probe's stiffness *x,y,z* components were set to 3 N m^−1^, 3 N m^−1^ and 50 N m^−1^, respectively.
Data availability {#Sec10}
-----------------
X-Ray crystallographic data can be obtained free of charge from the Cambridge Crystallographic Data Centre (CCDC) with the CCDC number 1529624 via its website (<https://www.ccdc.cam.ac.uk/structures/>). All other data are available within the article and its Supplementary Information or from the corresponding authors upon reasonable request.
Electronic supplementary material
=================================
{#Sec11}
Supplementary Information
**Electronic supplementary material**
**Supplementary Information** accompanies this paper at doi: 10.1038/s41467-017-01934-1.
**Publisher\'s note:** Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
C.-A.P. was supported from the European Union's Horizon 2020 research and innovation program 2D ink (no. 664878). X.-Y.W. is grateful to a fellowship from the Alexander von Humboldt Foundation. R.B. appreciates support by European Social Fund and the Federal State of Saxony (ESF-Project "GRAPHD"). M.G. acknowledges the H2020-MSCA-IF-2014 program and W.A. a Heisenberg professorship by the DFG. Dr Tilo Lübken is acknowledged for help in NMR measurements. This work was partially supported by the ERC Consolidator Grant NanoSurfs (no. 615233), the Max Planck Society, the German Excellence Initiative "Center for Advancing Electronics Dresden" (cfaed), and the Graphene Flagship.
X.-Y.W., M.R., Y.H., J.B. and A.R. contributed equally to this work. X.-Y.W., M.R., R.R. and R.B. performed the solution experiments and related calculations. Y.H., A.R., M.G. and C.-A.P. performed the experiments at surfaces. J.B. performed the DFT simulations at interfaces and C.-A.P. the AFM simulations. F.H. and J.J.W. characterized the crystals. W.A. and J.V.B. designed and conceived the UHV methods. A.N., R.B., X.F., C.-A.P. and K.M. conceived and designed the experiments. All the authors contributed to the manuscript.
Competing interests {#FPar1}
===================
The authors declare no competing financial interests.
|
{
"pile_set_name": "PubMed Central"
}
|
Seattle’s KEXP-FM won the Station of the Year Award at the recent CMJ music festival and convention in New York. It’s the second year in a row the nonprofit public music station has won the award to the annual music industry gathering sponsored by the College Music Journal.
[Read More]
Rock musician Thomas Dolby of “Blinded by Science” fame leads the list of music industry guests at the Recording Academy’s Music and Technology Summit Saturday from 9 a.m. to 6 p.m. at the JBL Theater and Learning Lab at Experience Music Project.
[Read More]
It’s hard to believe that the annual Seattle International Standup Comedy Competition started more than a generation ago. It’s now older than many of the comics who compete in the yearly marathon.
[Read More]
|
{
"pile_set_name": "Pile-CC"
}
|
S-R Media, The Spokesman-Review and Spokesman.com are happy to assist you. Contact Customer Service by email
or call 800-338-8801
WASHINGTON – The U.S. has often overestimated the ability of Afghan military and police units to fight on their own, according to an independent report released Monday that calls into question the strategy to win the war and bring troops home.
The investigation is the first objective look at the rating system the military has used for the past five years to judge the effectiveness of Afghan troops. Its findings seem to contradict upbeat assessments recently provided by senior military commanders overseeing the war.
The capability of Afghan forces is considered the single biggest indicator of whether the war is …
You have viewed 20 free articles or blogs allowed within a 30-day period. FREE registration is now required for uninterrupted access.
Registration Required
log in to your Spokesman.com account for unlimited viewing and commenting access.
S-R Media, The Spokesman-Review and Spokesman.com are happy to assist you. Contact Customer Service by email
or call 800-338-8801
WASHINGTON – The U.S. has often overestimated the ability of Afghan military and police units to fight on their own, according to an independent report released Monday that calls into question the strategy to win the war and bring troops home.
The investigation is the first objective look at the rating system the military has used for the past five years to judge the effectiveness of Afghan troops. Its findings seem to contradict upbeat assessments recently provided by senior military commanders overseeing the war.
The capability of Afghan forces is considered the single biggest indicator of whether the war is going well and is seen as the linchpin in the U.S. strategy since the war began more than eight years ago.
Lawmakers are likely to use the latest findings to question President Barack Obama’s handling of the war. Democrats say they are frustrated that Obama is sending more U.S. troops into combat without assurances that Afghan forces are close behind.
“The bottom line to this is that the system … is flawed, it’s unreliable and it’s inconsistent,” said Arnold Fields, who led the study as the special inspector general for Afghanistan reconstruction.
Two weeks before he was fired by Obama, Gen. Stanley McChrystal told reporters that “their growth is on track” and “we’re ahead of the plan.” But the report found that the system used to judge that success was deeply flawed. In some cases, units with the same rating would have different abilities. Also, highly rated units often regressed as soon as U.S. mentors withdrew.
In one stark example, a police district in the northern Afghan province of Baghlan was given the top rating by NATO officials in August 2008. The “CM1” designation meant the police were independently capable of conducting operations. But when investigators asked to visit the district in February, they were told the district wasn’t secure and was overrun with insurgents.
One official told investigators that the police force had “withered away to the point that it barely functions.”
In a written response to the report, NATO says it has made significant progress in both training Afghan forces and measuring their effectiveness in recent months.
Lt. Gen. Bill Caldwell, who heads the training mission in Afghanistan, said that NATO has suffered a severe shortage in trainers with some facilities on the verge of shutting down.
“Building an enduring and self-sustaining force remains a distinct challenge and attainment of the growth objectives is not assured,” Caldwell wrote in a letter to investigators.
|
{
"pile_set_name": "Pile-CC"
}
|
Temozolomide stability in extemporaneously compounded oral suspensions.
Temozolomide, commercially available in capsules, is an oral alkylating agent used to treat brain tumors. The purpose of this study was to determine the pharmaceutical acceptability and chemical stability of temozolomide in two extemporaneously compounded suspension formulations prepared from the capsules. The temozolomide oral suspensions were prepared from 100-mg commercial capsules yielding a nominal temozolomide concentration of 10 mg/mL. The suspension vehicles selected for testing were an equal parts mixture of Ora-Plus and Ora-Sweet and an equal parts mixture of Ora-Plus and Ora-Sweet SF. The suspensions were packaged in amber plastic screw-cap prescription bottles, which were stored at 23 deg C for 21 days or 4 deg C for 60 days. Stability-indicating high-performance liquid chromatographic analysis revealed that the temozolomide concentration in both suspension vehicle combinations exhibited little or no loss for 60 days at 4 deg C. At 23 deg C, temozolomide losses were somewhat greater. In the Ora-Sweet formulation, the loss was 6% at 7 days; in the Ora-Sweet SF formulation, losses were about 8% at 14 days and 10% to 11% at 21 days. Temozolomide extemporaneously prepared as oral suspensions from capsules in equal parts mixtures of Ora-Plus suspension vehicle with Ora-Sweet and with ora-Sweet SF syrups with added povidine k-30 and acidified with citric acid were pharmaceutically acceptable and chemically stable for at least 60 days at 4 deg C. Refrigerated storage is recommended. The suspensions should not be stored at room temperature longer than 1 week if Ora-Sweet is used or longer than 2 weeks if Ora-Sweet SF is used.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
The news of water droplets trickling from Jesus’ feet at a cross at Irla spread like wildfire on Monday, and attracted hundreds through the day to witness the ‘miracle.’
“I am a follower of Jesus, and the moment I heard of this, I came rushing with a bottle. I have taken three to four drops, since others in the line also needed to take blessings,” said Papu Deobandar, 35, homemaker from Nehru Nagar. She then stood waiting for her sister to help her get some ‘holy water.’
People brought vessels of different sizes to collect the water.
Some took pictures and videos, and compared it with previous miracles. “In Chakala, too, there was a similar miracle. This is basically Jesus’s way of helping us wash and cleanse our sins,” said Denzil Gonsalves, another resident.
Sharada Parmar, 62, was the first one to spot the ‘miracle,’ and said that she could not believe it. “After my home puja and Narayan’s puja, I had come to pray at the cross as usual at 8.30 am. While I was praying, water droplets fell on me. When it continued to fall, I was too shocked to believe it.”
Overwhelmed, she called her neighbour to witness it. “I checked His head, hands and shoulders. Everything was dry except the feet from where the water was dripping,” said Perpetual Fernandes, 53-year-old homemaker and resident of the area. “It was then confirmed that it was a miracle, and I collected some of the water to sprinkle at home.”
The cross, which has the statue, is next to Edward Dias wadi in Irla. “It was first built in 1873 as a thanksgiving for the first son born in the Dias family. He was my grandfather,” said Leander Dias, who called the event a miracle.
“The statue came up 10-12 years ago,” said Collen Gomes, a local.
“We reworked the cross in 2001, as it was in a bad condition. Some relics of the old were put into the new one, and the old one was buried in our compound,” said Gordon, another member of the Dias family.
|
{
"pile_set_name": "OpenWebText2"
}
|
Tenodesis of the extensor carpi ulnaris for chronic, post-traumatic lunotriquetral instability.
We describe a technique of soft-tissue reconstruction which is effective for the treatment of chronic lunotriquetral instability. Part of extensor carpi ulnaris is harvested with its distal attachment preserved. It is passed through two drill holes in the triquetrum and sutured to itself. This stabilises the ulnar side of the wrist. We have reviewed 46 patients who underwent this procedure for post-traumatic lunotriquetral instability with clinical signs suggestive of ulnar-sided carpal instability. Standard radiographs were normal. All patients had pre-operative arthroscopy of the wrist at which dynamic lunotriquetral instability was demonstrated. A clinical rating system for the wrist by the Mayo clinic was used to measure the outcome. In 19 patients the result was excellent, in ten good, in 11 satisfactory and in six poor. On questioning, 40 (87%) patients said that surgery had substantially improved the condition and that they would recommend the operation. However, six (13%) were unhappy with the outcome and would not undergo the procedure again for a similar problem. There were six complications, five of which related to pisotriquetral problems. The mean follow-up was 39.1 months (6 to 100). We believe that tenodesis of extensor carpi ulnaris is a very satisfactory procedure for isolated, chronic post-traumatic lunotriquetral instability in selected patients. In those with associated pathology, the symptoms were improved, but the results were less predictable.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Q:
The auto_increment column is still not continuous after rebuilding in MySQL table
There is a table named "web_file" with an auto_increment column "num_id" in Mysql (version 8.0) database. Because of some delete operations, the "num_id" is not continuous. So I drop the original "num_id", and add a new auto_increment column still named "num_id" started from 1. But I find the maximum value of the new "num_id" isn't equal with the count of the rows. Here is my sql code:
alter table web_file drop column num_id;
alter table web_file add column num_id int not null auto_increment unique key;
select min(num_id), max(num_id), count(num_id) from web_file;
This is the result:
1 664291 662902
I have tried to execute the sql code several times, but the result are all the same every time. In my opinion, when a column is auto_increment continuously, its max value should be equal with the number of rows. So, what's wrong leading to this exception?
A:
What you are seeing is not really an exception so much as it is just the way auto increment columns in databases behave. The contract for an auto increment column is that when a new id gets added, it sequentially follows the previous greatest value, and that all id values are guaranteed to be unique. There is no guarantee that the sequence will be continuous.
If you require a continuous sequence, then you may use the ROW_NUMBER analytic function (MySQL 8+ and later):
SELECT *, ROW_NUMBER() OVER (ORDER BY num_id) rn
FROM web_file;
On earlier version of MySQL, you could use a correlated subquery:
SELECT *,
(SELECT COUNT(*) FROM web_file w2 WHERE t2.num_id <= t1.num_id) rn
FROM web_file t1;
|
{
"pile_set_name": "StackExchange"
}
|
There was a typographical error in the legend of Figure 7. Different fonts were used.
**Competing Interests:**No competing interests declared.
|
{
"pile_set_name": "PubMed Central"
}
|
Diagnosis coding in The Danish National Patient Registry for respiratory syncytial virus infections.
This national registry-based epidemiological study aimed to evaluate the usefulness of the Danish National Patient Registry (DNPR) as a tool for epidemiological studies of respiratory syncytial virus (RSV) hospitalizations. Information on RSV diagnoses in records of hospitalizations among inpatients tested for RSV in Denmark from January 1996 to May 2003 in the DNPR was validated against the RSV test result in records from the 18 laboratories testing for RSV among hospitalized patients in Denmark. Of 16,733 RSV-positive samples representing a total of 14,898 hospitalizations in the DNPR, 68% (10,111) had been registered in the DNPR with 1 or more RSV diagnoses. Age influenced the odds ratio (OR) of being coded with a RSV diagnosis among patients with a RSV positive test, hence 73% of infants compared to 57% of patients beyond infancy had been registered in the DNPR with 1 or more RSV diagnoses. The OR of being coded with a RSV diagnosis among patients with an RSV positive test was increased over the study period and was also increased by increasing length of hospitalization and by the presence of secondary diagnoses. The OR was decreased by the presence of asthma and other chronic conditions in the patients. The OR was furthermore influenced by RSV seasonality and by hospital. Taking into account a loss of 27% of RSV hospitalizations, DNPR can be used for epidemiological studies of RSV among infants in Denmark.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Q:
Helper Class Error - Grouping in Aggregate
I don't get an error in the console, but for the below helper class but when I edit my record I get this error on save:
"Error:Apex trigger RoomTypeTrigger caused an unexpected exception, contact your administrator: RoomTypeTrigger: execution of AfterUpdate caused by: System.UnexpectedException: field 'Room_Description__c' can not be grouped in a query call: Class.AccountRoomRollups.rollupLineItems: line 26, column 1"
I understand I am grouping in an aggregate and that might not be acceptable, I know I am missing something but can't figure out what. NOTE: This is my very first trigger!!!
USE CASE: the trigger worked fine with just the SUM updates to account, but then I added the room description grouping because I need to also pull the Room Description field for all children when "Need image" = TRUE on the child and enter each description in one text field on the parent account (a list of all room descriptions for children that need an image).
public without sharing class AccountRoomRollups {
public static void rollupLineItems(List<Room_Type__c> lineItems) {
if (lineItems != null && lineItems.isEmpty() == false) {
//Get Account Ids
Set<Id> accountIds = new Set<Id>();
for (Room_Type__c lineItem : lineItems) {
if (lineItem.Hotel_Account__c != null){
accountIds.add(lineItem.Hotel_Account__c);
}
}
//Get Map of Account
Map<Id, Account> Accounts = new Map<Id, Account>([SELECT Id, Room_Images_Mapped_Sum__c,
Rooms_with_Image_Mapped_Count__c,
Room_Types_Count__c,Rooms_without_Images__c
FROM Account WHERE Id IN :accountIds]);
//Aggregate Room_Images_Mapped_Sum__c
Map<Id, Double> mappedSum = new Map<Id, Double>();
Map<Id, Double> mappedRmTrue = new Map<Id, Double>();
Map<Id, Double> countRm = new Map<Id, Double>();
//List<String> Description = new List<String>();
String Descriptionstring = '';
AggregateResult[] results = [SELECT Hotel_Account__c, SUM(Image_Count_mapped__c) RollupAmount,
SUM(Mapped_Images_True__c) MappedTrue,
SUM(Room_Type_Count__c) RCount, Room_Description__c,Needs_a_mapped_image__c
FROM Room_Type__c WHERE Hotel_Account__c IN :accountIds
GROUP BY Hotel_Account__c,Room_Description__c,Needs_a_mapped_image__c];
for (AggregateResult result : results) {
Id parentId = (Id) result.get('Hotel_Account__c');
Double rollupAmount = (Double) result.get('RollupAmount');
Double mappedTrue = (Double) result.get('MappedTrue');
Double rCount = (Double) result.get('RCount');
mappedSum.put(parentId, rollupAmount);
mappedRmTrue.put(parentId, mappedTrue);
countRm.put(parentId, rCount);
Boolean needimage = (Boolean) result.get('Needs_a_mapped_image__c');
String roomdescription = (String) result.get('Room_Description__c');
IF (needimage = True) {
Descriptionstring += roomdescription + '\n';
}
}
//Map Amounts for Update
List<Account> accToUpdate = new List<Account>();
for(Id parentId : accounts.keySet()) {
Account acc = accounts.get(parentId);
Double rollupAmount = 0;
Double mappedTrue = 0;
Double rCount = 0;
if (mappedSum.containsKey(parentId)||mappedRmTrue.containsKey(parentId)) {
rollupAmount = mappedSum.get(parentId);
mappedTrue = mappedRmTrue.get(parentId);
rCount = countRm.get(parentId);
}
if (rollupAmount != acc.Room_Images_Mapped_Sum__c || mappedTrue != acc.Rooms_with_Image_Mapped_Count__c || rCount != acc.Room_Types_Count__c){
acc.Room_Images_Mapped_Sum__c = rollupAmount;
acc.Rooms_with_Image_Mapped_Count__c = mappedTrue;
acc.Room_Types_Count__c = rCount;
acc.Rooms_without_Images__c = Descriptionstring;
accToUpdate.add(acc);
}
}
//Update Account
if (accToUpdate.isEmpty() == false) {
update accToUpdate;
}
}
}
}
A:
You cannot GROUP BY a Text Area field, which it looks like Room_Description__c is. If you really must group by description, you may want to populate the value of this field into a plain Text field somewhere, or just change the type to Text. Text fields can be grouped.
|
{
"pile_set_name": "StackExchange"
}
|
Patch Problems? Characteristics of Transdermal Drug Delivery System Exposures Reported to the National Poison Data System.
Transdermal drug delivery systems (TDDS) pose special risks considering a large amount of drug they contain and their modified release properties. We sought to characterize TDDS exposures reported to the National Poison Data System (NPDS). NPDS was searched for all human exposures to a TDDS from 1/1/2006 to 12/31/2015. Only single-substance TDDS exposures followed to a known medical outcome were included for final analysis. Specific data analyzed was date, sex, age, TDDS product, exposure reason, route of exposure, medical outcome, management site, level of health care facility care, clinical effects, and interventions. Over that 10-year period, 6746 adults and 1917 pediatric exposures were identified. Exposures declined over the study period. The most common exposure reason in adults was intentional abuse (n = 1622) compared to unintentional-general (n =1 070) in pediatric cases. TDDS ingestion was reported in 4519 adults and 2825 pediatric cases. Fentanyl was the most common substance encountered in adult (n = 4656) and pediatric cases (n = 474). No or minor effect were the most common medical outcomes in both groups. In fentanyl cases, moderate or major outcomes were seen in 54 % (n = 1062) of adult and 26 % (n = 54) of pediatric cases. Naloxone was given in 1080 cases. Ninety-seven deaths (91 adults, 6 pediatrics) were reported, all involving ingestion of the TDDS. Fentanyl was associated with 80 adult and 5 pediatric deaths. Overall, single-substance TDDS exposures decreased over the duration of this study and typically resulted in no or mild effects. However, exposures involving fentanyl resulted in higher rates of major or moderate medical outcomes and were associated with multiple deaths.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
//========= Copyright Valve Corporation, All rights reserved. ============//
//
// Purpose:
//
//=============================================================================//
#ifndef GAMERULES_REGISTER_H
#define GAMERULES_REGISTER_H
#ifdef _WIN32
#pragma once
#endif
// Each game rules class must register using this in it's .cpp file.
#if !defined(_STATIC_LINKED)
#define REGISTER_GAMERULES_CLASS( className ) \
void __CreateGameRules_##className() { new className; } \
static CGameRulesRegister __g_GameRulesRegister_##className( #className, __CreateGameRules_##className );
#else
#define REGISTER_GAMERULES_CLASS( className ) \
void MAKE_NAME_UNIQUE(__CreateGameRules_)##className() { new className; } \
static CGameRulesRegister __g_GameRulesRegister_##className( #className, MAKE_NAME_UNIQUE(__CreateGameRules_)##className );
#endif
#ifdef _XBOX
// force symbol expansion
#define REGISTER_GAMERULES_CLASS2( className ) REGISTER_GAMERULES_CLASS( className )
#endif
class CGameRulesRegister
{
public:
typedef void (*CreateGameRulesFn)();
CGameRulesRegister( const char *pClassName, CreateGameRulesFn fn );
// Allocates the gamerules object associated with this class.
void CreateGameRules();
static CGameRulesRegister* FindByName( const char *pName );
private:
const char *m_pClassName;
CreateGameRulesFn m_pFn;
CGameRulesRegister *m_pNext; // Links it into the global list.
static CGameRulesRegister *s_pHead;
};
#ifdef CLIENT_DLL
// The client forwards this call so the game rules manager can create the appropriate
// game rules class.
void InstallStringTableCallback_GameRules();
#else
// Server calls this at startup.
void CreateNetworkStringTables_GameRules();
// Server calls this to install a specific game rules object. The class should have been registered
// with REGISTER_GAMERULES_CLASS.
void CreateGameRulesObject( const char *pClassName );
#endif
#endif // GAMERULES_REGISTER_H
|
{
"pile_set_name": "Github"
}
|
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<!--NewPage-->
<HTML>
<HEAD>
<!-- Generated by javadoc (build 1.5.0_11) on Mon Jan 19 00:08:31 CET 2009 -->
<TITLE>
GetMapsAction
</TITLE>
<META NAME="keywords" CONTENT="it.eng.spagobi.engines.geo.xservice.GetMapsAction class">
<LINK REL ="stylesheet" TYPE="text/css" HREF="../../../../../../stylesheet.css" TITLE="Style">
<SCRIPT type="text/javascript">
function windowTitle()
{
parent.document.title="GetMapsAction";
}
</SCRIPT>
<NOSCRIPT>
</NOSCRIPT>
</HEAD>
<BODY BGCOLOR="white" onload="windowTitle();">
<!-- ========= START OF TOP NAVBAR ======= -->
<A NAME="navbar_top"><!-- --></A>
<A HREF="#skip-navbar_top" title="Skip navigation links"></A>
<TABLE BORDER="0" WIDTH="100%" CELLPADDING="1" CELLSPACING="0" SUMMARY="">
<TR>
<TD COLSPAN=2 BGCOLOR="#EEEEFF" CLASS="NavBarCell1">
<A NAME="navbar_top_firstrow"><!-- --></A>
<TABLE BORDER="0" CELLPADDING="0" CELLSPACING="3" SUMMARY="">
<TR ALIGN="center" VALIGN="top">
<TD BGCOLOR="#EEEEFF" CLASS="NavBarCell1"> <A HREF="../../../../../../overview-summary.html"><FONT CLASS="NavBarFont1"><B>Overview</B></FONT></A> </TD>
<TD BGCOLOR="#EEEEFF" CLASS="NavBarCell1"> <A HREF="package-summary.html"><FONT CLASS="NavBarFont1"><B>Package</B></FONT></A> </TD>
<TD BGCOLOR="#FFFFFF" CLASS="NavBarCell1Rev"> <FONT CLASS="NavBarFont1Rev"><B>Class</B></FONT> </TD>
<TD BGCOLOR="#EEEEFF" CLASS="NavBarCell1"> <A HREF="class-use/GetMapsAction.html"><FONT CLASS="NavBarFont1"><B>Use</B></FONT></A> </TD>
<TD BGCOLOR="#EEEEFF" CLASS="NavBarCell1"> <A HREF="package-tree.html"><FONT CLASS="NavBarFont1"><B>Tree</B></FONT></A> </TD>
<TD BGCOLOR="#EEEEFF" CLASS="NavBarCell1"> <A HREF="../../../../../../deprecated-list.html"><FONT CLASS="NavBarFont1"><B>Deprecated</B></FONT></A> </TD>
<TD BGCOLOR="#EEEEFF" CLASS="NavBarCell1"> <A HREF="../../../../../../index-all.html"><FONT CLASS="NavBarFont1"><B>Index</B></FONT></A> </TD>
<TD BGCOLOR="#EEEEFF" CLASS="NavBarCell1"> <A HREF="../../../../../../help-doc.html"><FONT CLASS="NavBarFont1"><B>Help</B></FONT></A> </TD>
</TR>
</TABLE>
</TD>
<TD ALIGN="right" VALIGN="top" ROWSPAN=3><EM>
</EM>
</TD>
</TR>
<TR>
<TD BGCOLOR="white" CLASS="NavBarCell2"><FONT SIZE="-2">
<A HREF="../../../../../../it/eng/spagobi/engines/geo/xservice/GetHierarchiesAction.html" title="class in it.eng.spagobi.engines.geo.xservice"><B>PREV CLASS</B></A>
<A HREF="../../../../../../it/eng/spagobi/engines/geo/xservice/SaveAnalysisStateAction.html" title="class in it.eng.spagobi.engines.geo.xservice"><B>NEXT CLASS</B></A></FONT></TD>
<TD BGCOLOR="white" CLASS="NavBarCell2"><FONT SIZE="-2">
<A HREF="../../../../../../index.html?it/eng/spagobi/engines/geo/xservice/GetMapsAction.html" target="_top"><B>FRAMES</B></A>
<A HREF="GetMapsAction.html" target="_top"><B>NO FRAMES</B></A>
<SCRIPT type="text/javascript">
<!--
if(window==top) {
document.writeln('<A HREF="../../../../../../allclasses-noframe.html"><B>All Classes</B></A>');
}
//-->
</SCRIPT>
<NOSCRIPT>
<A HREF="../../../../../../allclasses-noframe.html"><B>All Classes</B></A>
</NOSCRIPT>
</FONT></TD>
</TR>
<TR>
<TD VALIGN="top" CLASS="NavBarCell3"><FONT SIZE="-2">
SUMMARY: NESTED | <A HREF="#field_summary">FIELD</A> | <A HREF="#constructor_summary">CONSTR</A> | <A HREF="#method_summary">METHOD</A></FONT></TD>
<TD VALIGN="top" CLASS="NavBarCell3"><FONT SIZE="-2">
DETAIL: <A HREF="#field_detail">FIELD</A> | <A HREF="#constructor_detail">CONSTR</A> | <A HREF="#method_detail">METHOD</A></FONT></TD>
</TR>
</TABLE>
<A NAME="skip-navbar_top"></A>
<!-- ========= END OF TOP NAVBAR ========= -->
<HR>
<!-- ======== START OF CLASS DATA ======== -->
<H2>
<FONT SIZE="-1">
it.eng.spagobi.engines.geo.xservice</FONT>
<BR>
Class GetMapsAction</H2>
<PRE>
java.lang.Object
<IMG SRC="../../../../../../resources/inherit.gif" ALT="extended by ">AbstractEngineAction
<IMG SRC="../../../../../../resources/inherit.gif" ALT="extended by "><A HREF="../../../../../../it/eng/spagobi/engines/geo/commons/service/AbstractGeoEngineAction.html" title="class in it.eng.spagobi.engines.geo.commons.service">it.eng.spagobi.engines.geo.commons.service.AbstractGeoEngineAction</A>
<IMG SRC="../../../../../../resources/inherit.gif" ALT="extended by "><B>it.eng.spagobi.engines.geo.xservice.GetMapsAction</B>
</PRE>
<HR>
<DL>
<DT><PRE>public class <B>GetMapsAction</B><DT>extends <A HREF="../../../../../../it/eng/spagobi/engines/geo/commons/service/AbstractGeoEngineAction.html" title="class in it.eng.spagobi.engines.geo.commons.service">AbstractGeoEngineAction</A></DL>
</PRE>
<P>
The Class GetMapsAction.
<P>
<P>
<DL>
<DT><B>Author:</B></DT>
<DD>Andrea Gioia ([email protected])</DD>
</DL>
<HR>
<P>
<!-- =========== FIELD SUMMARY =========== -->
<A NAME="field_summary"><!-- --></A>
<TABLE BORDER="1" WIDTH="100%" CELLPADDING="3" CELLSPACING="0" SUMMARY="">
<TR BGCOLOR="#CCCCFF" CLASS="TableHeadingColor">
<TH ALIGN="left" COLSPAN="2"><FONT SIZE="+2">
<B>Field Summary</B></FONT></TH>
</TR>
<TR BGCOLOR="white" CLASS="TableRowColor">
<TD ALIGN="right" VALIGN="top" WIDTH="1%"><FONT SIZE="-1">
<CODE>static java.lang.String</CODE></FONT></TD>
<TD><CODE><B><A HREF="../../../../../../it/eng/spagobi/engines/geo/xservice/GetMapsAction.html#FEATURE_NAME">FEATURE_NAME</A></B></CODE>
<BR>
</TD>
</TR>
</TABLE>
<!-- ======== CONSTRUCTOR SUMMARY ======== -->
<A NAME="constructor_summary"><!-- --></A>
<TABLE BORDER="1" WIDTH="100%" CELLPADDING="3" CELLSPACING="0" SUMMARY="">
<TR BGCOLOR="#CCCCFF" CLASS="TableHeadingColor">
<TH ALIGN="left" COLSPAN="2"><FONT SIZE="+2">
<B>Constructor Summary</B></FONT></TH>
</TR>
<TR BGCOLOR="white" CLASS="TableRowColor">
<TD><CODE><B><A HREF="../../../../../../it/eng/spagobi/engines/geo/xservice/GetMapsAction.html#GetMapsAction()">GetMapsAction</A></B>()</CODE>
<BR>
</TD>
</TR>
</TABLE>
<!-- ========== METHOD SUMMARY =========== -->
<A NAME="method_summary"><!-- --></A>
<TABLE BORDER="1" WIDTH="100%" CELLPADDING="3" CELLSPACING="0" SUMMARY="">
<TR BGCOLOR="#CCCCFF" CLASS="TableHeadingColor">
<TH ALIGN="left" COLSPAN="2"><FONT SIZE="+2">
<B>Method Summary</B></FONT></TH>
</TR>
<TR BGCOLOR="white" CLASS="TableRowColor">
<TD ALIGN="right" VALIGN="top" WIDTH="1%"><FONT SIZE="-1">
<CODE> void</CODE></FONT></TD>
<TD><CODE><B><A HREF="../../../../../../it/eng/spagobi/engines/geo/xservice/GetMapsAction.html#service(it.eng.spago.base.SourceBean, it.eng.spago.base.SourceBean)">service</A></B>(it.eng.spago.base.SourceBean request,
it.eng.spago.base.SourceBean response)</CODE>
<BR>
</TD>
</TR>
</TABLE>
<A NAME="methods_inherited_from_class_it.eng.spagobi.engines.geo.commons.service.AbstractGeoEngineAction"><!-- --></A>
<TABLE BORDER="1" WIDTH="100%" CELLPADDING="3" CELLSPACING="0" SUMMARY="">
<TR BGCOLOR="#EEEEFF" CLASS="TableSubHeadingColor">
<TH ALIGN="left"><B>Methods inherited from class it.eng.spagobi.engines.geo.commons.service.<A HREF="../../../../../../it/eng/spagobi/engines/geo/commons/service/AbstractGeoEngineAction.html" title="class in it.eng.spagobi.engines.geo.commons.service">AbstractGeoEngineAction</A></B></TH>
</TR>
<TR BGCOLOR="white" CLASS="TableRowColor">
<TD><CODE><A HREF="../../../../../../it/eng/spagobi/engines/geo/commons/service/AbstractGeoEngineAction.html#getGeoEngineInstance()">getGeoEngineInstance</A></CODE></TD>
</TR>
</TABLE>
<A NAME="methods_inherited_from_class_java.lang.Object"><!-- --></A>
<TABLE BORDER="1" WIDTH="100%" CELLPADDING="3" CELLSPACING="0" SUMMARY="">
<TR BGCOLOR="#EEEEFF" CLASS="TableSubHeadingColor">
<TH ALIGN="left"><B>Methods inherited from class java.lang.Object</B></TH>
</TR>
<TR BGCOLOR="white" CLASS="TableRowColor">
<TD><CODE>clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait</CODE></TD>
</TR>
</TABLE>
<P>
<!-- ============ FIELD DETAIL =========== -->
<A NAME="field_detail"><!-- --></A>
<TABLE BORDER="1" WIDTH="100%" CELLPADDING="3" CELLSPACING="0" SUMMARY="">
<TR BGCOLOR="#CCCCFF" CLASS="TableHeadingColor">
<TH ALIGN="left" COLSPAN="1"><FONT SIZE="+2">
<B>Field Detail</B></FONT></TH>
</TR>
</TABLE>
<A NAME="FEATURE_NAME"><!-- --></A><H3>
FEATURE_NAME</H3>
<PRE>
public static final java.lang.String <B>FEATURE_NAME</B></PRE>
<DL>
<DL>
<DT><B>See Also:</B><DD><A HREF="../../../../../../constant-values.html#it.eng.spagobi.engines.geo.xservice.GetMapsAction.FEATURE_NAME">Constant Field Values</A></DL>
</DL>
<!-- ========= CONSTRUCTOR DETAIL ======== -->
<A NAME="constructor_detail"><!-- --></A>
<TABLE BORDER="1" WIDTH="100%" CELLPADDING="3" CELLSPACING="0" SUMMARY="">
<TR BGCOLOR="#CCCCFF" CLASS="TableHeadingColor">
<TH ALIGN="left" COLSPAN="1"><FONT SIZE="+2">
<B>Constructor Detail</B></FONT></TH>
</TR>
</TABLE>
<A NAME="GetMapsAction()"><!-- --></A><H3>
GetMapsAction</H3>
<PRE>
public <B>GetMapsAction</B>()</PRE>
<DL>
</DL>
<!-- ============ METHOD DETAIL ========== -->
<A NAME="method_detail"><!-- --></A>
<TABLE BORDER="1" WIDTH="100%" CELLPADDING="3" CELLSPACING="0" SUMMARY="">
<TR BGCOLOR="#CCCCFF" CLASS="TableHeadingColor">
<TH ALIGN="left" COLSPAN="1"><FONT SIZE="+2">
<B>Method Detail</B></FONT></TH>
</TR>
</TABLE>
<A NAME="service(it.eng.spago.base.SourceBean, it.eng.spago.base.SourceBean)"><!-- --></A><H3>
service</H3>
<PRE>
public void <B>service</B>(it.eng.spago.base.SourceBean request,
it.eng.spago.base.SourceBean response)</PRE>
<DL>
<DD><DL>
</DL>
</DD>
</DL>
<!-- ========= END OF CLASS DATA ========= -->
<HR>
<!-- ======= START OF BOTTOM NAVBAR ====== -->
<A NAME="navbar_bottom"><!-- --></A>
<A HREF="#skip-navbar_bottom" title="Skip navigation links"></A>
<TABLE BORDER="0" WIDTH="100%" CELLPADDING="1" CELLSPACING="0" SUMMARY="">
<TR>
<TD COLSPAN=2 BGCOLOR="#EEEEFF" CLASS="NavBarCell1">
<A NAME="navbar_bottom_firstrow"><!-- --></A>
<TABLE BORDER="0" CELLPADDING="0" CELLSPACING="3" SUMMARY="">
<TR ALIGN="center" VALIGN="top">
<TD BGCOLOR="#EEEEFF" CLASS="NavBarCell1"> <A HREF="../../../../../../overview-summary.html"><FONT CLASS="NavBarFont1"><B>Overview</B></FONT></A> </TD>
<TD BGCOLOR="#EEEEFF" CLASS="NavBarCell1"> <A HREF="package-summary.html"><FONT CLASS="NavBarFont1"><B>Package</B></FONT></A> </TD>
<TD BGCOLOR="#FFFFFF" CLASS="NavBarCell1Rev"> <FONT CLASS="NavBarFont1Rev"><B>Class</B></FONT> </TD>
<TD BGCOLOR="#EEEEFF" CLASS="NavBarCell1"> <A HREF="class-use/GetMapsAction.html"><FONT CLASS="NavBarFont1"><B>Use</B></FONT></A> </TD>
<TD BGCOLOR="#EEEEFF" CLASS="NavBarCell1"> <A HREF="package-tree.html"><FONT CLASS="NavBarFont1"><B>Tree</B></FONT></A> </TD>
<TD BGCOLOR="#EEEEFF" CLASS="NavBarCell1"> <A HREF="../../../../../../deprecated-list.html"><FONT CLASS="NavBarFont1"><B>Deprecated</B></FONT></A> </TD>
<TD BGCOLOR="#EEEEFF" CLASS="NavBarCell1"> <A HREF="../../../../../../index-all.html"><FONT CLASS="NavBarFont1"><B>Index</B></FONT></A> </TD>
<TD BGCOLOR="#EEEEFF" CLASS="NavBarCell1"> <A HREF="../../../../../../help-doc.html"><FONT CLASS="NavBarFont1"><B>Help</B></FONT></A> </TD>
</TR>
</TABLE>
</TD>
<TD ALIGN="right" VALIGN="top" ROWSPAN=3><EM>
</EM>
</TD>
</TR>
<TR>
<TD BGCOLOR="white" CLASS="NavBarCell2"><FONT SIZE="-2">
<A HREF="../../../../../../it/eng/spagobi/engines/geo/xservice/GetHierarchiesAction.html" title="class in it.eng.spagobi.engines.geo.xservice"><B>PREV CLASS</B></A>
<A HREF="../../../../../../it/eng/spagobi/engines/geo/xservice/SaveAnalysisStateAction.html" title="class in it.eng.spagobi.engines.geo.xservice"><B>NEXT CLASS</B></A></FONT></TD>
<TD BGCOLOR="white" CLASS="NavBarCell2"><FONT SIZE="-2">
<A HREF="../../../../../../index.html?it/eng/spagobi/engines/geo/xservice/GetMapsAction.html" target="_top"><B>FRAMES</B></A>
<A HREF="GetMapsAction.html" target="_top"><B>NO FRAMES</B></A>
<SCRIPT type="text/javascript">
<!--
if(window==top) {
document.writeln('<A HREF="../../../../../../allclasses-noframe.html"><B>All Classes</B></A>');
}
//-->
</SCRIPT>
<NOSCRIPT>
<A HREF="../../../../../../allclasses-noframe.html"><B>All Classes</B></A>
</NOSCRIPT>
</FONT></TD>
</TR>
<TR>
<TD VALIGN="top" CLASS="NavBarCell3"><FONT SIZE="-2">
SUMMARY: NESTED | <A HREF="#field_summary">FIELD</A> | <A HREF="#constructor_summary">CONSTR</A> | <A HREF="#method_summary">METHOD</A></FONT></TD>
<TD VALIGN="top" CLASS="NavBarCell3"><FONT SIZE="-2">
DETAIL: <A HREF="#field_detail">FIELD</A> | <A HREF="#constructor_detail">CONSTR</A> | <A HREF="#method_detail">METHOD</A></FONT></TD>
</TR>
</TABLE>
<A NAME="skip-navbar_bottom"></A>
<!-- ======== END OF BOTTOM NAVBAR ======= -->
<HR>
</BODY>
</HTML>
|
{
"pile_set_name": "Github"
}
|
The man who took down the Texas church gunman
55-year-old Stephen Willeford is being heralded by police as a 'Texas Hero.'
Author:
Mike James, USA TODAY , TEGNA
Published:
10:41 AM EST November 7, 2017
One minute he was relaxing barefoot at his home. In the next, Stephen Willeford stood behind a pickup truck, firing the shots that led to the demise of the gunman who massacred 26 people in a church.
And Willeford still had no shoes on.
The 55-year-old plumber and experienced rifleman is heralded as a hero by local police here. Kelley fatally shot 26 people and had just exited First Baptist Church when Willeford surprised him.
"I was scared for my own family, who live a half a block away," Willeford said, describing his instinct to run out of his house after his daughter told him someone was shooting up the church.
"I kept hearing the shots, one after another, very rapid shots - just 'pop pop pop pop' and I knew every one of those shots represented someone, that it was aimed at someone, that they weren't just random shots," Willeford told 40/29 News TV.
Willeford came to a vigil Monday night in honor of the victims. He was the center of attention.
More than a dozen members of the still-stunned community enveloped him in an emotional group hug that almost resembled a huddle at a pickup football game as religious music filled the air.
“He is the nicest guy you ever want to meet,” said Ron Leonard, Willeford’s cousin who has been at his side almost ever since the gunfire stopped. “He’d give you the shirt off his back. In fact, he’s kind of a little bit embarrassed by all this.”
Earlier, Leonard gently pushed back news cameras when they realized Willeford, absent from public view since the massacre, had joined more than 100 residents at the town’s athletic complex.
Willeford said he had very little time to think Sunday when his daughter told him about the shooting. He loaded his magazine and ran across the street to the church, not even taking the time to put on shoes. When Willeford saw the gunman, he exchanged gunfire.
"He saw me and I saw him," Willeford said in the TV interview. "I was standing behind a pickup truck for cover."
At least one of the shots that Willeford fired struck Kelley through the protective body armor that the killer was wearing.
"I know I hit him," Willeford said. "He got into his vehicle, and he fired another couple rounds through his side window. When the window dropped, I fired another round at him again."
The gunman then sped down the highway.
Willeford spotted a pickup truck at a stop sign. He ran to the truck and asked the driver for help.
"That guy just shot up the Baptist church. We need to stop him," Willeford told the driver.
Willeford had help from another local resident, Johnnie Langendorff, who said he was driving past the church as the shooting happened. The armed man asked to get in Langendorff's truck, and the pair followed as the gunman drove away.
"He jumped in my truck and said, 'He just shot up the church. We need to go get him.' And I said 'Let's go,'" Langendorff said.
The pursuit reached speeds up to 90 mph. The gunman eventually lost control of his vehicle and crashed. The armed man walked up to the vehicle with his gun drawn, and the attacker did not move. Police arrived about five minutes later, Langendorff said.
The assailant was dead in his vehicle. He had three gunshot wounds — two from where the armed man hit him in the leg and the torso and the third self-inflicted wound to the head, authorities said.
"There was no thinking about it. There was just doing. That was the key to all this. Act now. Ask questions later," Langendorff said.
Willeford said he's no hero.
"I think my God, my Lord protected me and gave me the skills to do what needed to be done," Willeford said.
Willeford's family has lived in the Sutherland Springs area for four generations, 40/29 News reported. He had numerous friends who went to the church.
|
{
"pile_set_name": "Pile-CC"
}
|
RBC
: red blood cell
|
{
"pile_set_name": "PubMed Central"
}
|
1. Field of the Invention
The present invention relates to microprocessors and, in particular, to several aspects of a data processing system architecture that efficiently integrates a low-cost multiply/accumulate unit.
2. Discussion of the Prior Art
As shown in FIG. 1, a microprogrammed integrated circuit data processing system 10 includes a central processing unit 12 to manipulate data in accordance with operating software that comprises a set of program instructions. Data and program instructions utilized by the CPU 12 are stored in an associated memory 14. Transfer of instructions and data between the CPU 12 and the memory 14 is implemented by a bus interface mechanism which can be part of the CPU 12. Communication between the system 10 and other systems or peripheral devices is accomplished via an input/output device 16.
The memory 14 is typically organized in words, each containing N bits of information, i.e., instructions or data, and each having an address that specifies its location in the memory 14. The CPU 12 retrieves (reads) or provides (writes) information a word at a time by addressing a storage location in the memory 14 and either reading the word stored in that location or writing new information into that location.
Since accessing memory 14 is a slow process relative to the data processing speed of the CPU 12, as shown in FIG. 2, the CPU 12 usually includes an internal register file 18 comprising a number of its own registers which can be read or written very quickly. Since the register file 18 is internal to the CPU 12, many operations can be carried out by the CPU 12 without accessing the memory 14.
As further shown in FIG. 2, the CPU 12 also includes an arithmetic logic unit (ALU) 20 that actually performs the data manipulations specified by the program instructions. The ALU 20 usually receives two operands from the register file 18 via a multiplexor MUX and provides a single result at its output. In some CPU architectures, one of the input operands is always stored in a special internal accumulator register and the result of an ALU operation is always written into this accumulator register.
The CPU 12 also typically includes a shifter 22 for shifting the contents of an internal register, or as shown in FIG. 2, the output of the ALU 20, one or more bits in either direction to provide multiply and divide capability.
Thus, the CPU 12 includes all of the elements necessary to perform all arithmetic and logical data manipulations specified by a program instruction.
To implement a flow of program instructions, that is, to execute the microprocessor's operating program, the CPU 12 relies on a program counter (PC) 24 and its associated control logic to retrieve a series of program instructions and associated data from memory 14. The program counter 24 may simply increment itself through a sequence of program instruction addresses or modify the normal flow of program instructions by responding to special conditions which cause the program counter to "jump" or "branch" to instruction subroutines that depend on the special condition.
In a microprogrammed processor, the actual execution of a particular program instruction is accomplished by performing a specific sequence of microinstructions. Each microinstruction provides the control signals needed to set the ALU 20 to perform a corresponding "micro-operation" and specifies the next microinstruction in the microinstruction sequence for that program instruction.
In some applications, the processor's instruction set includes a program instruction that initiates a microinstruction sequence for performing a series of repetitive math operations to sample or condition data. For example, in so-called digital signal processing (DSP) applications, the processor recovers digital data from a modulated analog input signal utilizing a filtering technique that includes iterative multiply and accumulate steps based on the number of "taps" included in the filter.
A typical DSP multiply-accumulate microinstruction sequence is shown in FIG. 3. First, data is fetched from a sample buffer. Next, filter tap coefficient data is fetched from a coefficient memory. The retrieved data sample (multiplier) and the coefficient (multiplicand) are then multiplied and the resulting product term is added to an accumulating register. The memory pointers are then incremented to repeat the procedure for each filter tap, with product terms being accumulated throughout the procedure to provide a final filter output. The faster the filter throughput, the faster the data can be recovered.
In one prior art approach, the Texas Instruments TI320 DSP Microprocessor, a very fast multiply/accumulate instruction is incorporated into the processor's instruction set. This instruction allows the processor to fetch data, perform the math operations and manage the sample/coefficient memory fast enough to permit practical use of the TI320 processor in DSP applications.
A recognized improvement to the Texas Instruments approach is to add a circular buffer manager unit to the processor. This speeds up the microprocessor by off-loading responsibility for data management.
A circular buffer can be implemented in a number of ways. One popular technique utilizes a set of registers pointing to corresponding locations in memory. One pointer points to the active filter tap sample; this is a dynamic pointer that sequences to the next sample after each multiply operation. A second pointer points to the top of the sample stack. A third pointer points to the bottom of the stack. The circular buffer manager keeps track of the active sample pointer. When the active sample pointer matches the third pointer, it is reloaded to the top of the stack.
It would, however, be desirable to have available a microprocessor that implements a multiply/accumulate function with as few clock cycles as possible for a reasonable hardware cost.
|
{
"pile_set_name": "USPTO Backgrounds"
}
|
Rahul Gandhi recently concluded a two-day visit to his Wayanad constituency
Congress leader Rahul Gandhi Tuesday wrote to Prime Minister Narendra Modi, seeking his intervention to protect both the life and livelihood of the people in his Lok Sabha constituency Wayanad, ravaged by floods and landslides.
In his letter, the Wayanad MP suggested that many lives can be saved if early warning systems are installed to alert people living in vulnerable areas in the ecologically sensitive hill district.
Rahul Gandhi, who concluded his two-day visit to his flood-hit constituency on Monday, explained to the Prime Minister the issues being faced by people and ecology in Wayanad and suggested measures for his consideration.
"Developing a long term strategy and an action plan supported through a special package of assistance to protect both the life and livelihood of people and environment in Wayanad to be taken up on priority," Mr Gandhi wrote.
"Many lives can be saved if early warning systems are installed to alert populations in vulnerable areas and specially designed landslide/flood shelters with communication facilities are provided in such places," he said.
He said the destruction of forests and the depletion of inflow into the streams and rivers endanger farms and populations in three states- Kerala, Karnataka and Tamil Nadu.
"Mining and quarrying continues in the Western Ghats. The depletion of the forests is rapidly increasing conflicts between animals and people living within and in the periphery of the forest," he said.
Mr Gandhi said Wayanad has some of the most fragile parts of the Western Ghats and is home to 10 per cent of the biodiversity of the country.
"I would like to highlight the twin crisis one that affects the life and livelihood of people in Wayanad and the second, the critical status of the ecology in an area that is known as a global hotspot for biodiversity with the potential to contribute to carbon sequestration benefiting the country as a whole," he wrote.
Highlighting the crisis in the agricultural sector in Wayanad, Rahul Gandhi said a large number of farmers have committed suicide in the last two decades due to crisis in the farm sector.
He said agriculture and animal husbandry are the mainstays of the people in Wayanad.
However, paddy cultivation that also sustained wetlands and has the capacity to hold excess rainwater and mitigate flood damage, has been steadily declining since 1970s, he added.
Rahul Gandhi had tweeted on Friday that he spoke to PM Modi on the flood situation in Kerala and sought assistance for those affected in the State, including in his constituency Wayanad.
|
{
"pile_set_name": "OpenWebText2"
}
|
Please check my quantum physics browser game for accuracy - sideshowb
https://linkingideasblog.wordpress.com/2016/04/25/learning-quantum-mechanics-the-easy-way/
======
evanb
[1/4] Source: I am a nuclear physicist.
Full disclosure: I didn't play. I'm not launching a Java applet. So I'm just
reading your description for physics content.
> (The independence of quantum systems from absolute phase is called gauge
> symmetry)
Gauge symmetry isn't the independence of the global phase---it's the
independence of a spacetime-dependent phase. This is a much _much_ bigger
symmetry than the independence of global absolute phase.
> Until that question is answered, then we can’t rule out the possibility that
> consciousness does have something to do with it.
Are you sure you only read Feynman? This is an extreme fringe position. I
don't know any physicists who actually holds this position, unless they're
trying to get on TV.
> Interestingly, the Schrodinger equation states that a particle can’t exist
> at all in a place with more potential than the particle has energy. It will
> simply “jump” down any available holes to satisfy the equation. The same
> jumping behaviour is what gives rise to tunelling, when a particle jumps as
> if by magic from one hole to another.
That's exactly the opposite of what the Schrodinger equation says! A particle
_can_ exist in a place where the potential energy is more than the total
energy. That cannot happen classically, but is allowed quantum-mechanically.
However, the wavefunction dies off exponentially fast. But it's never 0 unless
the potential is infinite. That's what allows tunneling---the exponentially
small tail is real.
There is no "jumping" in quantum mechanics. The Schrodinger equation implies
the continuity equation [continuity].
~~~
Steuard
Not that a quality post like this needs it, but for what it's worth, as a
physics professor I'd happily endorse just about everything you've said in
this detailed response to the text accompanying the game (all four parts). (My
own comments focused mostly on the game itself, to the extent that I get
what's going on under the hood.)
~~~
dxbydt
Aside: Hey you are the guy who wrote those awesome tutorials on lagrange
multipliers!!! I actually used your materials multiple times at official
presentations, devtimes and other occasions where the management and
developers wanted to know how bidding optimization worked under the hood.
Thank you so much for writing so well.
~~~
Steuard
I'm glad to hear that you've appreciated that tutorial! It's especially neat
to hear about it being useful to people who aren't just cramming for a class.
------
alexbock
You may get more feedback if you upload the Java source to GitHub and make it
easy to compile and run locally. I wasn't able to run the applet with the
default settings in Safari, Chrome, nor Firefox and asking people to make
their browsers less secure to test this is probably going to turn a lot of
people off.
~~~
virgil_disgr4ce
Agreed, this is SUPER disappointing. I started reading your blog post on the
train this morning and was SO excited, because I've also been self-studying
QM/QED lately, and your description so closely mirrored my experiences. So
imagine my excitement when I saw that you'd made a game as a way of developing
an intuition—a really fantastic idea, and one I had been thinking of myself!
And yet I have effectively no way of even seeing it! Not even the code?!?!
:((((
~~~
brianwawok
So its annoying enough for you to type a 300 word complaint, but not annoying
enough to change your browser configuration? Only on hacker news I guess...
~~~
Vendan
Well, on my end, it would involve installing another browser and changing it's
config. Not doing that, thanks.
~~~
virgil_disgr4ce
Exactly.
------
valarauca1
Web Java is functionally dead outside of enterprises environments. Running
untrusted Java in browser is a huge security risk.
If you'd upload your code to github that'd make reviewing it easier.
~~~
earlz
Yea, I'm not taking that risk.. much less the inconvenience of installing a
different browser and getting Java enabled
------
kecks
Sorry, but I'm not running your Java applet.
Perhaps you could port your game to something like Processing.js? It allows
for very Java-like syntax, compiles to html5/js and is good with graphics.
~~~
RobSis
Unless you have a problem with running unknown compiled code, why not?
It's as easy as running '$ appletviewer URL'
~~~
whamlastxmas
Because having Java enabled for a browser is tremendously more risky than
running a single application of unknown code.
~~~
LukeShu
As RobSis noted, you don't have to enable it in your browser, you can use
appletviewer to run individual applets without the browser:
$ appletviewer http://tropic.org.uk/~crispin/quantum/
And also, moderns browsers allow you to have the plugin installed, but
disabled unless you activate it on a specific page.
~~~
kecks
Modern browsers... except Google Chrome, which has at least 40% market share.
Google Chrome does not support applets in any way since the deprecation of
NPAPI.
------
leni536
I'm reading through your description and this catched my eye: "In general, the
smaller the wave, the faster it spreads out: this is the uncertainty
principle." This is not true, spreading out is due to dispersion. Why you
don't see particle like wave packets on water surface? Also dispersion. In
fact you can create particle like waves on a guitar string, which is pretty
much dispersionless. Also you can describe massless particles in QM which are
also dispersionless.
The uncertanity principle is about the uncertanity of non-commuting
observables at any given time (eg. position and momentum), it's not about the
time dependence of the wave function.
~~~
Steuard
It's fairly common for people to relate this quantum phenomenon to the
uncertainty principle, though. Someone asks: "Why can't I just say 'the
particle is at rest in this small area' and expect it to stay there forever?"
And one take on the answer is "Because you're imposing a very small
uncertainty on the initial position, the initial momentum must have a very
high uncertainty. That means there's a substantial probability that the
particle will move away from its initial position over time, which corresponds
to the wave function spreading out."
You can certainly talk about it in terms of Fourier components or that sort of
thing, too. There's a degree to which the uncertainty principle as applied to
the position/momentum degrees of freedom is nothing _but_ a statement about
Fourier transforms. But I don't think it's a bad conceptual shorthand,
especially for students who aren't yet experts in thinking about Fourier
stuff.
~~~
tamana
As a non Fourier expert, thinking about the general concept of the Fourier
transform (spatial vs frequency coordinates),and how position and momentum are
Fourier transforms of each other, was highly illuminating
------
coldpie
Crashes when run in IcedTea 1.6 in Firefox on Linux.
IcedTea-Web Plugin version: 1.6.2 6/1/16 7:49 AM Exception was:
net.sourceforge.jnlp.LaunchException: Fatal: Initialization Error: Could not
initialize applet. For more information click "more information button". at
net.sourceforge.jnlp.Launcher.createApplet(Launcher.java:764) at
net.sourceforge.jnlp.Launcher.getApplet(Launcher.java:686) at
net.sourceforge.jnlp.Launcher$TgThread.run(Launcher.java:933) Caused by:
java.lang.ClassNotFoundException: quantum.GameWindow at
net.sourceforge.jnlp.runtime.JNLPClassLoader.loadClass(JNLPClassLoader.java:1562)
at net.sourceforge.jnlp.Launcher.createApplet(Launcher.java:753) ... 2 more
This is the list of exceptions that occurred launching your applet. Please
note, those exceptions can originate from multiple applets. For a helpful bug
report, be sure to run only one applet. 1) at 6/1/16 7:49 AM
net.sourceforge.jnlp.LaunchException: Fatal: Initialization Error: Could not
initialize applet. For more information click "more information button". at
net.sourceforge.jnlp.Launcher.createApplet(Launcher.java:764) at
net.sourceforge.jnlp.Launcher.getApplet(Launcher.java:686) at
net.sourceforge.jnlp.Launcher$TgThread.run(Launcher.java:933) Caused by:
java.lang.ClassNotFoundException: quantum.GameWindow at
net.sourceforge.jnlp.runtime.JNLPClassLoader.loadClass(JNLPClassLoader.java:1562)
at net.sourceforge.jnlp.Launcher.createApplet(Launcher.java:753) ... 2 more
~~~
LukeShu
Same exception with IcedTea 1.6 in Iceweasel on GNU/Linux.
Second edit: It's because I had OpenJDK 7 as the default JRE. It works fine
when I switch it to OpenJDK 8.
Tip: stick 4 spaces before the stack trace to not munge newlines.
Edit: If I run it with appletviewer, I get a more helpful exception:
$ appletviewer http://tropic.org.uk/~crispin/quantum/
java.lang.UnsupportedClassVersionError: quantum/GameWindow : Unsupported major.minor version 52.0
at java.lang.ClassLoader.defineClass1(Native Method)
[snip]
~~~
mschulze
This error means that you need at least JRE8 (Java 8) to run the applet. Or OP
must compile the applet in a compatible bytecode format for older versions of
Java (-target of javac).
~~~
LukeShu
Yup, I figured that out. Note that what we saw in the browser didn't indicate
that; I had to use appletviewer to get a helpful error message.
------
stared
For the game - I would be delighted to test it, but most likely this weekend
or so. (Plus, Java sadly isn't the most user-friendly interface for browser-
based things.)
For getting feedback from quantum information researchers, I recommend posting
to
[https://www.facebook.com/groups/qinfo.scientists.unite/](https://www.facebook.com/groups/qinfo.scientists.unite/).
I see that you have similar taste/inspirations to mine (I am also a big fan of
Test Tube Games (Velocity Raptor & Agent Higgs)). See also:
[https://hackpad.com/J0X4MSberlM?r=0](https://hackpad.com/J0X4MSberlM?r=0) for
more picks (and feel invited to add more).
For quantum games - in a week or so I plan to release my own -
[https://github.com/stared/quantum-game](https://github.com/stared/quantum-
game)
~~~
sideshowb
Thanks. I'll check out the FB group you mention. Your game, or what we can
currently see of it, reminds me of the (classical) "Reflections" laser game of
a few years back which is a good thing! And I suspect the hackpad list will
take me more than a weekend to play through :)
~~~
stared
I've heard about Reflections but it was mostly inspired by Chromatron
([http://silverspaceship.com/chromatron/](http://silverspaceship.com/chromatron/)),
and a bit by The Incredible Machine.
With this list - don't try to play it all at once. Some of these are potent
drugs - not only addictive, but mind-altering (e.g. Hyperrogue, when you walk
on a hyperbolic plane).
------
benbenolson
Yeah, I don't think that many people are going to be able to run this. Java
applets are deprecated at this point, unfortunately. I'd love to be able to
compile and run it locally, though.
~~~
jeffwass
SideShowB - since you implemented these as Java applets and as per your link
want to learn more about QM, and specifically about improving learning and
visualisation of QM, check out the below link.
It's a bunch of QM applets I created about 12 years ago to aid in visualising
different aspects of quantum mechanics. I was a grad student at the time, and
made these as part of a fellowship. Several of these were used in
corresponding homework assignments and lectures for both grad and undergrad QM
classes.
[http://www.pha.jhu.edu/~javalab/](http://www.pha.jhu.edu/~javalab/)
I decided to apply for this fellowship after an enlightening experience seeing
the time evolution of Coherent States of a quantum simple harmonic oscillator.
------
Coding_Cat
To me (coffee-deprived Physics Bsc.) nothing jumps out as particularly
erroneous, pretty fun. Save for some weird behaviour when you try to move the
particles to fast as is noted in the game. I could also get some parts of the
wave function outside of the bounds of the level (but within the applet) I'm
not sure if this is supposed to happen? How do you model/simulate the fields?
I was actually thinking of making my own little Quantum game to see if it
could be made fun. I was thinking of making a QGolf system, where you tweak an
initial stationary wavefunction which is then collapsed after a certain time.
Although these kinds of mechanics probably work better.
>Until I made this game I was pretty confused about the uncertainty principle.
The way it’s usually taught, without resorting to maths, is to say something
like this:
>Until I made Quantum Marble Maze (QMM) I had never heard a satisfying
explanation for why frequency and momentum are the same thing. This is called
Planck’s relation. Sure, you can demonstrate classically that higher frequency
waves carry more energy, but does that really mean a particle changes
frequency if you change its momentum – in other words if you push it?
If you know how to do Fourier Transforms, note that the momentum-wavefunction
is the Fourier Transform of the postion-wavefunction. And in order to make a
well defined peak using sines & cosines, you need a lot of high frequency
(high momentum) waves. IIRC this is shown in Griffiths Introduction to Quantum
Mechanics, quite a good introduction to QM.
~~~
sideshowb
Thanks. On your latter point, I do get Fourier transforms, though for that
logic to apply you first need to accept frequency-as-momentum :)
------
LionessLover
I think you should try
[http://physics.stackexchange.com/](http://physics.stackexchange.com/)
~~~
sideshowb
I did. Well, I didn't ask the main QA site as questions about specific sites
would be off-topic. I did ask chat (the H Bar) but didn't get much of a
response. An hour on HN and my server log says several people have loaded the
game and one made it to level 5, so we're doing better here already.
~~~
sideshowb
...3 hours in, 174 people have loaded the game and 5 made it to the final
(16th) level. Still no completions though ;)
On the other hand, most of the comments have been about the Java rather than
the physics. I suppose I should have expected that.
~~~
tgb
I read your blog and thought it was coherent (ha) and correct as far as I can
tell with one course qm and one on quantum computing plus one other textbook
as my background. Honestly one of the better descriptions of QM I've read
though it's hard for me to judge how someone without any background would take
to it.
------
gpsx
I wish I could play the game. I have the same java applet problem. Hopefully
I'll be able to play later.
A few words on quantum mechanics - it is very tricky. Someone doesn't
understand quantum mechanics by taking a class in it or in getting an
undergraduate physics degree. I am sure even many (or all) physics professors
have limitations in their knowledge. It's not that the rules are complicated,
it is the implications when it is applied in the real world. You do seem to
have gained a good level of understanding though.
I have a comment on the nature of a wave function, as described in the section
"It's wavelike, but not watery". It states that wavefuntions can act like
particles. This may be wording that I just do not understand, but I would want
to clarify it.
Take the wave function for an electron - this is not a wave of "electon". It
is a wave of probability (sort of) for an electron. The electron itself is
always a particle. The motion of the electron is described by the wave. In
this way the electron has wave properties.
To take an example, let's shoot an electron gun at a target. Suppose the
resulting wavefunction of the electron has a uniform amplitude over the target
when it "hits". What is the damage to the plate? Is it uniform damage? No. The
damage is always at a single point of impact. We see the result of the
electron hitting the target, not the wave function hitting the target.
Now, if we shoot lots of electrons, the damage will be pretty uniform, since
they will be distributed all over the plate. But it will still consist of a
bunch of point impacts.
I just wanted to be clear about differentiating the wave function from the
object whose state (position, momentum, whatever) is described by the wave
function.
------
crottypeter
The following (while not wrong) is hard to read. You lament that people think
"a particle does have definite position and momentum" but don't go on to state
that they do _not_ have definite x and p. Why not state this unambiguously?
> This can lead to a bit of confusion, because a lot of people take that to
> mean that a particle does have definite position and momentum (I mean, all
> things do, right?!) but you’re sadly not able to find out what it is. Maybe
> because trying to measure one changes the other – logically that makes
> sense, right?
> But this is QM, so until you start turning it into computer games, it won’t
> make sense
------
dluan
This is one of the coolest things I've ever seen.
------
mcguire
It's a Java applet?
~~~
sideshowb
Alas yes! In hindsight not the best choice of tech so a Javascript rewrite
might be on the cards.
~~~
Filligree
Kotlin has a JS compilation mode, and IDEA has a java-to-kotlin converter.
Maybe that's a good place to start? Kotlin is a good language anyway.
~~~
lmm
Scala has both those things and is a better and more mature language.
~~~
mushishi
It is a different language to be sure but why would you say it is better?
Isn't there such a many ways to look at the strengths/weaknesses of a language
that simply valuing one language over another does not make much sense? (In
this context, I assume that both are good enough to be possible for this
project).
~~~
lmm
More consistent design - a lot of things that are special cases in Kotlin are
just the logical combination of two orthogonal features in Scala (or Ceylon).
That and the huge power of higher-kinded types - there are some language
features where it's arguable whether they're worth their weight, but higher-
kinded types are one where it's really clear-cut (IMO) - once you're used to
them, using a language without them is like using a language without generics.
For the sake of argument I'll accept that it's maybe difficult to compare two
languages in general, but in terms of Kotlin/Scala they are extremely similar
languages targeting extremely similar use cases.
~~~
softawre
We reviewed both and went with Kotlin. Scala is not the obvious choice here in
my opinion.
Kotlin is a simpler language which is a huge benefit.
[https://kotlinlang.org/docs/reference/comparison-to-
scala.ht...](https://kotlinlang.org/docs/reference/comparison-to-scala.html)
~~~
lmm
It's not a simpler language. It's a less powerful one but as a result it has a
lot of ad-hoc special case features to cover subsets of functionality that
Scala provides with a few powerful features applied consistently.
(If you want an actually simpler language, look at Ceylon)
------
kqr
On a completely different spin: any tips for becoming good at the game? I
can't even pass level two, and I've already exhausted my (very limited) mental
QM book.
~~~
sideshowb
Most levels are easier to solve if you don't hang about: the longer you wait,
the more uncertainty bites and the position becomes indefinite.
Remember there's no friction so if you accelerate over half the screen you
need to decelerate over the other half if you're going to stop before the
other side.
Level 2 can be solved by accelerating/decelerating to follow three straight
lines (east-south-west) to the goal. Alternatively if you're feeling cocky you
can accelerate east then throw in some south as you approach the corner:
you'll bounce off the walls and break the wavefunction a bit but about 50% of
the time the goal will trigger.
If you haven't discovered already you can also skip levels from the menu ;)
~~~
kqr
Thanks, that helped a lot! I also felt really clever when I realised that if
there's a "Collapse 5%" trigger zone I barely have to do anything other than
nudge the wave in the right direction when some parts of it are near the
collapse trigger. =)
------
capnrefsmmat
This reminds me of the old Quantum Minigolf game, though that was desktop C++:
[http://quantumminigolf.sourceforge.net/](http://quantumminigolf.sourceforge.net/)
Instead of arrow keys, you use a putter to tap the quantum golf ball.
~~~
sideshowb
I'm starting to think if I had made this a downloadable C++ executable I'd
have fewer complaints about security :D
------
Steuard
A few thoughts, speaking as a physics professor who asks his relativity
students to play both "Velocity Raptor" and "A Slower Speed of Light":
* Like many others here, I was a bit dismayed to see this implemented as a Java applet. Increasing security concerns have meant that I've been able to use fewer and fewer of those in my classes: it's just not reasonable to ask students to jump through those hoops anymore.
* When I look at the menu, it only goes up through level 7.
* I have made very little progress on level 8 (the one with the diffraction grating in the middle that's trying to get you to reach 2% transmission at about an 80 degree angle on each side). I have a general idea of what you want me to do, but I get so little visible transmission most of the time that it feels all but impossible to fine-tune my momentum or other aspects of my strategy. (Is there any way to increase the slit widths or number of slits or something to make that easier?)
* What _exactly_ do the arrow keys _do_? Are you imposing a linear potential gradient across the whole screen for a short time? It's hard to judge "accuracy" when I'm not quite sure what I'm looking at. (That also means I'm not entirely sure how to teach it.)
* Along similar lines, do you have a sense of what exactly is leading to the visible interference effects during motion? (The ones that you point out can be guides as to which direction the wave is moving?) Maybe I ought to have a direct intuition for this! Is it purely an effect of reflections off of the walls, or would I see the same thing if I started with a Gaussian wave packet in a linear potential without any boundaries at all? (And if it's reflection related, why is it mostly showing the direction of primary motion rather than the direction of the reflections?) Some explanation for this might be nice in a teaching tool.
* Do you think you could allow the user to toggle the phase-rainbow mode on and off? (That's the mode that I _would_ expect to always give an indication of the dominant momentum of the wave function.)
* The "collapse" mechanic is very helpful for game play, but might deserve some sort of explanation. In particular, I worry that it could train students to believe a critical misconception: that if the probability of a particular state becomes high enough, that state immediately becomes "true". Maybe that wouldn't turn out to be a serious concern, but I'm not certain that beginners are well-equipped to recognize which weird aspects of game play are supposed to reflect real physics and which are mere game enhancements. (The way collapse works here feels a lot like Copenhagen, but not quite, since it doesn't obey the Born probability rule.)
* It would allow for significantly richer strategy and game play if the player had a wider menu of potential functions to choose from. (For instance, what if the arrow keys continued to be linear potential functions and a mouse click provided a harmonic oscillator?)
* In your explanatory text, I'm not comfortable with your discussion of decoherence. I'm far from an expert on that subject, so I'm hesitant to try to offer specific corrections. But it feels off to me. (In particular, a single quantum particle bouncing off of strict potential energy walls should _not_ be a manifestation of decoherence, because the particle's state isn't becoming _entangled_ with the state of those walls in any way. I don't see any manifestation of decoherence in the game.)
In case you're interested, my current favorite intro level textbook on quantum
mechanics is Tom Moore's "Six Ideas that Shaped Physics: Unit Q". It doesn't
go as far as I'd like for a really thorough course (which is especially true
in the recent 3rd edition, though the material that's still there is presented
much more clearly), but as part of an introductory sequence it does a great
job. Much like Townsend's fantastic Jr/Sr level text, Moore first introduces
the "rules" of quantum mechanics in the context of two-state spin systems (in
a slightly simplified version of Dirac's bra-ket notation), and only moves to
wave functions once the fundamental concepts are established. (It doesn't
manage to do a whole lot with time dependent position/momentum systems like
the ones you're simulating, though.)
~~~
sideshowb
Another mega comment: don't know why this isn't more upvoted. Thanks for the
bug report.
Level 8 is odd. You thought more diffraction would occur with slower speed
right? In the simulation it doesn't; bounce off the left wall and hit the
grating fast to complete. I have to admit this bothers me. I can only assume
this is because sidebands which would otherwise hit the grating itself move
inwards to hit the 2% targets or something. Not good at all from a learning
point of view.
Yes, the arrow keys impose a linear potential gradient while they are being
held. It fades in gradually as well so no sharp transients (unless you
differentiate). Not sure if that detail was required but it's there now.
The interference effects during motion are not wall reflections as far as I
can tell. Some levels such as the traps don't have walls but (imperfect)
absorbers and they exhibit it just as much. You're right, it's easy to see why
this occurs in the rainbow mode but not so much in the amplitude mode. Genuine
physics or artifact of simulation - presumably somebody can do some calculus
and work that one out!
Decoherence, I acknowledge should involve multiple quantum particles. What we
do see here though is an evolution from a single particle with coherent phase,
which interferes with itself, into one with incoherent phase, that doesn't.
Conceptually that seems similar to decoherence.
I think all your pedagogical points are good ones btw.
Like I said to evanb, if you would like to be acknowledged for your review I'm
more than happy to. Best wishes.
~~~
Steuard
I'm very puzzled about level 8, based on what you've said. Certainly to be a
good _pedagogical_ tool, I'd want to see a system that very visibly showed a
wider and wider diffraction pattern for longer wavelengths (as you've said).
[The one advantage of a bounce that I was able to think of while playing was
that it might wind up giving a cleaner "plane wave" input, but I'm not even
sure that that's an advantage. Now that you mention it, though, perhaps there
is some ideal speed for which bouncing off the back _and_ side walls would
result in propagation at an angle, so that a meaningful fraction of the
probability density would hit the targets in as a _central_ interference
maximum.]
If I were you, I'd leave out the discussion of decoherence entirely. It's a
subtle topic (evanb did it justice, I think), and I'm reasonably convinced
that it doesn't really apply to the game. (Even if I might be wrong about
that, is that tidbit of background info important enough to your discussion to
be worth the effort and/or the risk?)
Thanks for a fun take on quantum stuff, regardless! (I don't know what's in
the higher levels, but might be fun to see some more examples of things like
tunneling through different barriers. And I wonder if there's any way of
setting up resonance in a cavity in an interesting way...)
Regarding credit: I don't feel like I've contributed _that_ much at this
point, but if you wind up with a moderate list of thank you's at some point,
you're welcome to include "Steuard Jensen" on the list.
[As for my comment not being more upvoted: I think I got some sort of black
mark on my record here a few months ago after I tried to insist that recent
reports about reactionless propulsion systems are very unlikely to pan out.
Ever since then, I've gotten the sense that my comments are showing up with an
initial scoring penalty.]
~~~
sideshowb
Re resonance in a cavity, I have some prototypes but they didn't make it into
gameplay. Actually the end screen is an example. Several levels do exhibit
resonance in a potential well: 7, 9, 11 and 12.
[Really? Wow, that's harsh. I haven't the foggiest how HN scores work, it's
all rather secretive isn't it.]
------
amelius
There are some physics people here, but... wouldn't it be a much better idea
to ask a physics forum?
------
em3rgent0rdr
Really enjoyed this game...I was very skeptical before I actually ran the
applet.
------
pvinis
Wow. I was about to write that I'm writing a simulation engine to try and
understand quantum physics better. Then I saw the image on your post, and I
now see that you are far ahead of me! :D
------
dingo_bat
Pretty cool game. Although I have been unable to map it to any of the limited
knowledge/understanding I possess about quantum mechanics.
------
javathrowaway
You asked for feedback, and here it goes: I couldn't run your applet on my
machine.
It's not my intent to attack the choice of Java, but I can't avoid noticing
that Java applets as a web technology are getting seriously obsolete these
days. It's a challenge to get these applets running.
JavaScript works way better on the web (I hate to say it, but beware the silly
name: JS as a programming language has nothing to do with Java).
The traceback follows.
Exception was:
net.sourceforge.jnlp.LaunchException: Fatal: Initialization Error: Could not initialize applet. For more information click "more information button".
at net.sourceforge.jnlp.Launcher.createApplet(Launcher.java:739)
at net.sourceforge.jnlp.Launcher.getApplet(Launcher.java:668)
at net.sourceforge.jnlp.Launcher$TgThread.run(Launcher.java:901)
Caused by: net.sourceforge.jnlp.LaunchException: The applet is signed but its manifest specifies Sandbox permissions. This is not yet supported. Try running the applet again, but choose the Sandbox run option.
at net.sourceforge.jnlp.runtime.ManifestAttributesChecker.checkPermissionsAttribute(ManifestAttributesChecker.java:217)
at net.sourceforge.jnlp.runtime.ManifestAttributesChecker.checkAll(ManifestAttributesChecker.java:82)
at net.sourceforge.jnlp.runtime.JNLPClassLoader. (JNLPClassLoader.java:288)
at net.sourceforge.jnlp.runtime.JNLPClassLoader.createInstance(JNLPClassLoader.java:351)
at net.sourceforge.jnlp.runtime.JNLPClassLoader.getInstance(JNLPClassLoader.java:418)
at net.sourceforge.jnlp.runtime.JNLPClassLoader.getInstance(JNLPClassLoader.java:394)
at net.sourceforge.jnlp.Launcher.createApplet(Launcher.java:704)
... 2 more
Caused by: net.sourceforge.jnlp.LaunchException: Fatal: Initialization Error: Run in Sandbox call performed too late. The classloader was notified to run the applet sandboxed, but security settings were already initialized.
at net.sourceforge.jnlp.runtime.JNLPClassLoader$SecurityDelegateImpl.setRunInSandbox(JNLPClassLoader.java:2386)
at net.sourceforge.jnlp.runtime.ManifestAttributesChecker.checkPermissionsAttribute(ManifestAttributesChecker.java:214)
... 8 more
This is the list of exceptions that occurred launching your applet. Please note, those exceptions can originate from multiple applets. For a helpful bug report, be sure to run only one applet.
1) at 6/1/16 7:09 PM
net.sourceforge.jnlp.LaunchException: Fatal: Initialization Error: Run in Sandbox call performed too late. The classloader was notified to run the applet sandboxed, but security settings were already initialized.
at net.sourceforge.jnlp.runtime.JNLPClassLoader$SecurityDelegateImpl.setRunInSandbox(JNLPClassLoader.java:2386)
at net.sourceforge.jnlp.runtime.ManifestAttributesChecker.checkPermissionsAttribute(ManifestAttributesChecker.java:214)
at net.sourceforge.jnlp.runtime.ManifestAttributesChecker.checkAll(ManifestAttributesChecker.java:82)
at net.sourceforge.jnlp.runtime.JNLPClassLoader. (JNLPClassLoader.java:288)
at net.sourceforge.jnlp.runtime.JNLPClassLoader.createInstance(JNLPClassLoader.java:351)
at net.sourceforge.jnlp.runtime.JNLPClassLoader.getInstance(JNLPClassLoader.java:418)
at net.sourceforge.jnlp.runtime.JNLPClassLoader.getInstance(JNLPClassLoader.java:394)
at net.sourceforge.jnlp.Launcher.createApplet(Launcher.java:704)
at net.sourceforge.jnlp.Launcher.getApplet(Launcher.java:668)
at net.sourceforge.jnlp.Launcher$TgThread.run(Launcher.java:901)
2) at 6/1/16 7:09 PM
net.sourceforge.jnlp.LaunchException: The applet is signed but its manifest specifies Sandbox permissions. This is not yet supported. Try running the applet again, but choose the Sandbox run option.
at net.sourceforge.jnlp.runtime.ManifestAttributesChecker.checkPermissionsAttribute(ManifestAttributesChecker.java:217)
at net.sourceforge.jnlp.runtime.ManifestAttributesChecker.checkAll(ManifestAttributesChecker.java:82)
at net.sourceforge.jnlp.runtime.JNLPClassLoader. (JNLPClassLoader.java:288)
at net.sourceforge.jnlp.runtime.JNLPClassLoader.createInstance(JNLPClassLoader.java:351)
at net.sourceforge.jnlp.runtime.JNLPClassLoader.getInstance(JNLPClassLoader.java:418)
at net.sourceforge.jnlp.runtime.JNLPClassLoader.getInstance(JNLPClassLoader.java:394)
at net.sourceforge.jnlp.Launcher.createApplet(Launcher.java:704)
at net.sourceforge.jnlp.Launcher.getApplet(Launcher.java:668)
at net.sourceforge.jnlp.Launcher$TgThread.run(Launcher.java:901)
Caused by: net.sourceforge.jnlp.LaunchException: Fatal: Initialization Error: Run in Sandbox call performed too late. The classloader was notified to run the applet sandboxed, but security settings were already initialized.
at net.sourceforge.jnlp.runtime.JNLPClassLoader$SecurityDelegateImpl.setRunInSandbox(JNLPClassLoader.java:2386)
at net.sourceforge.jnlp.runtime.ManifestAttributesChecker.checkPermissionsAttribute(ManifestAttributesChecker.java:214)
... 8 more
3) at 6/1/16 7:09 PM
net.sourceforge.jnlp.LaunchException: Fatal: Initialization Error: Could not initialize applet. For more information click "more information button".
at net.sourceforge.jnlp.Launcher.createApplet(Launcher.java:739)
at net.sourceforge.jnlp.Launcher.getApplet(Launcher.java:668)
at net.sourceforge.jnlp.Launcher$TgThread.run(Launcher.java:901)
Caused by: net.sourceforge.jnlp.LaunchException: The applet is signed but its manifest specifies Sandbox permissions. This is not yet supported. Try running the applet again, but choose the Sandbox run option.
at net.sourceforge.jnlp.runtime.ManifestAttributesChecker.checkPermissionsAttribute(ManifestAttributesChecker.java:217)
at net.sourceforge.jnlp.runtime.ManifestAttributesChecker.checkAll(ManifestAttributesChecker.java:82)
at net.sourceforge.jnlp.runtime.JNLPClassLoader. (JNLPClassLoader.java:288)
at net.sourceforge.jnlp.runtime.JNLPClassLoader.createInstance(JNLPClassLoader.java:351)
at net.sourceforge.jnlp.runtime.JNLPClassLoader.getInstance(JNLPClassLoader.java:418)
at net.sourceforge.jnlp.runtime.JNLPClassLoader.getInstance(JNLPClassLoader.java:394)
at net.sourceforge.jnlp.Launcher.createApplet(Launcher.java:704)
... 2 more
Caused by: net.sourceforge.jnlp.LaunchException: Fatal: Initialization Error: Run in Sandbox call performed too late. The classloader was notified to run the applet sandboxed, but security settings were already initialized.
at net.sourceforge.jnlp.runtime.JNLPClassLoader$SecurityDelegateImpl.setRunInSandbox(JNLPClassLoader.java:2386)
at net.sourceforge.jnlp.runtime.ManifestAttributesChecker.checkPermissionsAttribute(ManifestAttributesChecker.java:214)
... 8 more
|
{
"pile_set_name": "HackerNews"
}
|
A ex-presidente Dilma Rousseff lamentou no sábado 5 ter assinado a lei que prevê a colaboração premiada. “Infelizmente assinei a lei que criou a delação premiada. Digo infelizmente porque ela foi assinada genericamente, sem tipificação exaustiva. E a vida mostrou que sem tipificação exaustiva ela poderia virar uma arma de arbítrio, de absoluta exceção“, disse a ex-presidente.
Foi durante o primeiro mandato de Dilma, em 2013, que a colaboração premiada acabou sendo institucionalizada, por meio da sanção pela petista da Lei de Organizações Criminosas. Desde então, o instrumento tem sido uma ferramenta largamente utilizada pela força-tarefa da Operação Lava Jato, que sacudiu o mundo político e atingiu em cheio o PT, além do PMDB e do PP.
Dilma também comparou a forma como as delações premiadas da Lava Jato vêm sendo negociadas com uma forma de tortura. “Prendem e submetem a uma forma de controle. Na minha época era uma moleza: era só ser preso – já que torturavam, tinha pau de arara, choque, afogamento e morte.
Para nós era considerado leve, mas para as pessoas normais ser preso é gravíssimo: é perder a liberdade e o direito de ir e vir.
Por isso em alguns países se usa a exigência de que a delação só possa se dar sob condições voluntárias. Porque do contrário você submete e induz a pessoa a dizer o que você quer ouvir. A tortura faz isso. A tortura faz a pessoa dizer o que se quer. Às vezes a pessoa não diz, mas às vezes a pessoa mente. Se se mente sob tortura, imagina se não se mente sob delação premiada”, disse.
Leia também:
Dilma Rousseff: Compraram votos no impeachment e continuam a comprar
As declarações de Dilma foram feitas na abertura do Brazil Forum UK, na London School of Economics, em Londres. A conferência, que é organizada por estudantes brasileiros de várias universidades britânicas, tem como tema neste ano os 30 anos da Constituição de 1988. Mais cedo, o ministro do Supremo Tribunal Federal (STF) Luís Roberto Barroso também concedeu uma palestra.
Além da Lei das Organizações Criminosas, Dilma disse que seu governo e o de seu antecessor e padrinho político, o ex-presidente Luiz Inácio Lula da Silva, ajudaram a criar ou implementaram outros mecanismos contra a corrupção e o crime, como a nomeação de procuradores-gerais independentes e investimentos na Polícia Federal. “O lulopetismo era o inimigo a ser destruído. Utilizaram o que nós mesmos construímos contra nós”, disse.
“Lula é o candidato do PT”
Dilma foi recebida no fórum por uma ampla plateia de apoiadores. Vários membros do público gritaram frases em apoio a Lula e contra a Rede Globo e o juiz Sérgio Moro. Bem à vontade no ambiente, Dilma voltou a afirmar que foi vítima de um “golpe parlamentar”. Segundo ela, a condenação e prisão de Lula é mais uma etapa desse golpe. “Foi um golpe parlamentar. Articulado pelo partido da mídia e da toga, parte da elite e pelos partidos que organizaram”, disse ela.
Dilma também lançou outras críticas contra a Lava Jato. Ela afirmou que as cinco maiores empreiteiras do país foram “sistematicamente destruídas” durante o processo de combate à corrupção.
“Quando a Volkswagen e a Siemens foram pegas, ninguém destruiu as empresas”, disse Dilma, citando o caso das empresas alemãs que foram processadas por envolvimento em casos de falsificação de emissões e pagamentos de subornos. “Você teve um processo contra os corruptores, mas não contra a instituição.”
“Eram as cinco maiores empresas, que concorriam em qualquer lugar do mundo. Acho isso muito estranho”, disse, em referência às empreiteiras Odebrecht, Camargo Corrêa, OAS, Andrade Gutierrez e Queiroz Galvão.
Todas foram investigadas na Lava Jato e acusadas de participar de um esquema de corrupção que saqueou a Petrobras. “Você não destrói a instituição porque você destrói emprego”. Ela também criticou a Operação Carne Fraca, da Polícia Federal, que no ano passado mirou grandes frigoríficos do país por suspeita de irregularidades.
Por fim, Dilma voltou a afirmar que o PT não cogita outra candidatura à Presidência que não a de Lula, que está preso e é ameaçado pela Lei da Ficha Limpa.
“O PT não vai tirar o Lula nem oferecer outro candidato. Nós iremos sustentar a posição de inocência, e não cabe a nós tirar o Lula das eleições. Ele é uma ideia de unidade. Se o Lula participar, ele ganha a eleição”, disse Dilma.
Ela, no entanto, disse que não vê o PT dominando para sempre as candidaturas de esquerda no país em eleições posteriores a 2018. “Seria absurdo discutir plano B. Mas isso não significa que eu ache que os candidatos do futuro serão todos do PT. Nós vamos ter que passar o bastão para as novas gerações”, completou.
|
{
"pile_set_name": "OpenWebText2"
}
|
Introduction
============
The rapid progress of molecular cell biology and the constant need for improved drugs in medicine has created a need for structural biologists to look at increasingly more complex protein structures. These can often require particular post-translational modifications, molecular chaperones and co-factors to support their elaborate folding and enzymatic activity. Whilst the vast majority of structures present in the Protein Data Bank have been obtained using bacterial expression systems, prokaryotes are unable to perform a large number of these modifications and lack many essential eukaryotic co-factors. This can be an issue for the study of large multi-subunit complexes that are activated by small signaling molecules as well as for nuclear, cell surface and secreted proteins that require elaborate folding machineries. A number of *E. coli* strains have been engineered to overcome some of these limitations^1^. In recent years, however, the use of mammalian expression systems has been increasing because they reliably produce eukaryotic proteins that are otherwise problematic to express in other systems^2^. Today it is possible to obtain stable and transient mammalian expression cell lines through a variety of techniques that range from viral transduction to chemical-mediated transfection, physical gene-transfer such as electroporation and direct injection^3^. While all of these methods have their own set of advantages and disadvantages, only a few of them are suitable for structural studies being either too expensive and/or time consuming.
Here we describe a very simple, fast, inexpensive yet highly efficient method for expressing protein complexes for structural biology in suspension grown mammalian cells. The approach uses transient co-transfection of a Human Embryonic Kidney (HEK) cell line (*e.g.,* Freestyle HEK 293F cells). These cells have been derived from the HEK 293 cell line, are adapted to grow in suspension cultures, reaching high densities using serum-free media (such as FreeStyle 293 Expression Medium). Cells are then transiently transfected using a branched version of polyethylenimine (PEI), an inexpensive polymeric reagent that has been reported to function for a large range of mammalian cells^4^ by forming DNA/PEI complexes that enter the host cell by endocytosis^5^. This method is suitable for both small scale (30 ml) and large-scale (up to 300 ml) experiments and can produce high levels of purified protein complexes. It is particularly useful for studying proteins that require complex folding machineries, co-factors or particular post-translational modifications that cannot be performed by bacteria, yeast and insect cells.
In this protocol we present the expression and purification of the three-protein central scaffold from the Sin3A transcriptional repression complex. This consists of Histone Deacetylase 1 (HDAC1), Suppressor of Defective Silencing 3 (SDS3) and Switch-independent 3 (Sin3A). The purified complex is used for high-throughput crystallization trials.
Protocol
========
NOTE: The protocol is suitable for any scale of expression, therefore volumes and quantities of reagents should be scaled proportionally. A suitable mammalian expression vector must be used for this protocol. Here we used a modified pcDNA 3.1 expression vector that conveniently allows the inclusion or exclusion of an affinity tag depending on the choice of restriction enzymes used in the cloning (**Figure 1**). The following protocol describes a large scale transfection.
1. Large Scale Culture/Transfection in 1 L Roller Bottles
---------------------------------------------------------
Grow and maintain HEK293F suspension-adapted cells according to standard protocols. Typically starter cultures of between 30 and 100 ml are grown in 250 ml conical cell culture flasks.
1. Seed cells at 0.5 x 10^6^ cells/ml into a final volume of 300 ml in each 1 L roller bottle. NOTE: Roller bottles accommodate a minimum of 150 ml to a maximum of 300 ml of suspension culture. For larger scale transfections use multiple bottles.
2. Incubate for 24 hr in an orbital shaker incubator at 37 °C, 120 rpm, and 5% CO~2~ until cells reach a density of 1.0 x 10^6^ cells/ml (cells should divide approximately every 24 hr).
3. Pipette a total of 300 µg of filter-sterilized DNA (see supplementary method) into 30 ml of PBS and vortex vigorously for 3 sec. NOTE: Use a total of 1 µg of DNA per million transfected cells. If two or more plasmids are to be co-transfected reduce the amount of each so that the ratio of DNA to cells remains constant.
4. Add 1.2 ml of 0.5 mg/ml filter-sterilized PEI to the PBS/DNA solution and vortex vigorously for 3 sec.
5. Incubate the mix at RT for 20 min.
6. Add the DNA/PEI mix to the cells -- which should be at a density of 1 x 10^6^ cells/ml (step 1.3).
7. Following co-transfection, incubate the cells in an orbital shaker incubator for a further 48 hr at 37 °C, 120 rpm, and 5% CO~2~.
8. Harvest intracellular proteins by centrifuging cells at 3,000 x g for 5 min and store the pellet at -80 °C. NOTE: Alternatively, use a standard (non-CO~2~) shaking incubator, if the atmosphere above the culture is replaced with 5% CO~2~ and the bottle lid is sealed. The atmosphere will need to be replaced at each passage (normally every 2 days).
2. Protein-complex Purification from Whole Cell Extract
-------------------------------------------------------
This protocol is optimized for the purification of nuclear complexes using flag-tagged proteins.
1. Defrost cell pellet into \~40 ml of pre-chilled lysis buffer (100 mM potassium acetate, 50 mM Tris pH 7.5, 5% glycerol, 0.3% Triton X-100, protease inhibitors) per L of culture.
2. Re-suspend the pellet by pipetting up and down several times (to avoid foaming, do not vortex).
3. Thoroughly re-suspend cells using a glass homogenizer. Sonicate for 3 cycles (15 sec on, 15 sec off). Centrifuge at 30,000 x g for 25 min at 4ºC and retain supernatant.
4. Equilibrate 1.25 ml of anti-flag packed agarose resin per liter of culture by washing three times with resin equilibration buffer (100 mM potassium acetate, 50 mM Tris pH 7.5).
5. Incubate the whole cell extract from step 2.3 with the affinity resin in one or more 50 ml centrifuge tubes and gently rotate the sample for 30-120 min at 4 °C.
6. Centrifuge at 3,000 x g for 1 min at 4 °C, discard supernatant.
7. Wash the resin with 45 ml of pre-chilled buffer 1 (100 mM potassium acetate, 50 mM Tris pH 7.5, 5% glycerol, 0.3% Triton X-100). Centrifuge at 3,000 x g for 1 min and discard supernatant.
8. Repeat step 2.7 using high salt buffer (300 mM potassium acetate, 50 mM Tris pH 7.5, 5% glycerol), followed by low salt buffer (50 mM potassium acetate, 50 mM Tris pH 7.5, 5% glycerol) and TEV Cleavage Buffer (50 mM potassium acetate, 50 mM Tris pH 7.5, 0.5 mM TCEP). NOTE: Ensure each wash is brief*i.e.,* sufficient to fully re-suspend the resin and no longer.
9. Collect a 10 µl sample of resin and dilute into 1 volume of 2x protein loading buffer for analysis (bound protein control). Do not use reducing agents in order to avoid releasing the antibody from the resin.
10. Re-suspend the resin into 8-10 ml of pre-chilled TEV cleavage buffer and transfer to a 15 ml centrifuge tube.
11. Add \~40 µg of TEV protease (from stock at 1 mg/ml) per L of original culture and mix well by pipetting up and down several times.
12. Replace tube atmosphere with 100% N~2~ gas to prevent protein oxidation.
13. Gently rotate O/N at 4ºC.
14. Centrifuge at 3,000 x g for 10 min. Transfer the supernatant into an ultra centrifugal filter with an appropriate molecular weight cut-off and concentrate down to 500 µl.
15. Collect a 10 µl sample of the concentrated protein and dilute into 1 volume of 2x protein loading buffer for analysis. (TEV eluate control).
16. Collect a 10 µl sample of the resin and dilute into 1 volume of 2x protein loading buffer for analysis. Do not use reducing agents in order to avoid releasing large amounts antibody from the resin. (Post-TEV resin control).
17. Equilibrate the size exclusion chromatography column with gel filtration buffer (50 mM potassium acetate, 50 mM Tris pH 7.5, 0.5 mM TCEP).
18. Filter the protein through a 0.22 µm filter
19. Load the sample into the column and collect the fractions.
20. Run samples from steps 2.9, 2.15 and 2.16 and the gel filtration fractions from step 2.19 on an SDS-PAGE and Coomassie stain for analysis. NOTE: It is advisable to run an SDS-PAGE of samples from steps 2.9, 2.15 and 2.16 prior to gel filtration to confirm expression of the target protein.
Representative Results
======================
Here we show a 2 L (8 x 250 ml cultures) transient co-transfection and purification of the HDAC1, SDS3, Sin3A ternary complex. HDAC1 and SDS3 interact with Sin3A through the HDAC-interaction domain (HID) of Sin3A. A typical purification yields up to 1 mg of complex per liter of culture.
**Figure 1. Schematic of the modified pcDNA 3.1 expression vector.** The plasmid was digested with EcoRV and EcoRI to include the affinity-tag and TEV cleavage site on the amino terminus of Sin3A. To clone untagged versions of HDAC1 and SDS3, the vector was digested with KpnI and EcoRI.
**Figure 2. SDS-PAGE showing the first step of the purification.**The "bound protein" lane shows the complex bound to the affinity resin. Following TEV digestion, the tag present on Sin3A-HID is cleaved off and the complex is eluted from the resin as shown in the second and third lanes of the gel.
**Figure 3. Chromatogram of protein-complex purified using size exclusion chromatography.** Note that the pure complex eluted in the void volume because it forms a dimer in solution.
**Figure 4. SDS-PAGE showing the fractions of the gel filtration shown in Figure 3.**
**Figure 5. Trypan blue-stained HEK 293F cells at 2.3x10^6^cells/ml ready to be transfected.** Cells should only be present as single or dividing cells, while large clusters may be broken up by vigorous vortexing for approximately 25 sec.
------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------------------------------------------------------------------- ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
**Problem** **Possible causes** **Action**
Low protein yield. Low number of cells transfected. Make sure cell density is approximately 1.0 x 10^6^ before adding the transfection reaction mixture to the culture.
Cells might have been subcultured too many times. Use fresh stock cells after approximately 90 passages.
DNA may be degraded or have a high amount of impurities. Make sure the plasmid DNA being used has a 260/280 ratio between 1.8 and 2.0. Running the DNA on an agarose gel is advisable to assess its quality.
Protein(s) being expressed are not stable or soluble enough. Test different constructs in small scale prior to up-scaling the transfections. Make sure tags do not interfere with the structure of the protein(s) of interest.
Cells look cloudy and have an unusual color and/or odor. Cells are infected with bacteria or yeast. Good sterile technique must be used at all times. Fumigating the laminar flow hoods and UV-disinfecting the cell culture room in case of infection helps containing the problem.
Cells have low viability. Wrong pH in media. Make sure cultures are grown at 5-8% CO~2~at all times.
Cells have been cultured up to a density over 3.0x10^6^ cells/ml. Cells should not be grown up to densities above 2.5 x 10^6^cells/ml immediately before a transfection and never above 3.0 x 10^6^cells/ml.
Affinity purification did not work Protein(s) is not being expressed. See above.
Purification conditions might be wrong. Adjust buffer conditions (*e.g.,* high salt, low salt, pH.)
------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------------------------------------------------------------------- ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
**Table 1. Troubleshooting.**
-------------------------------------------- ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
**System** **Advantages** **Disadvantages**
***E. coli*** Fast expression systemHigh protein yieldsInexpensiveSuitable for the affordable production of isotopically labeled proteins Lack of important chaperones that may be required for protein foldingLack of many post translational modificationsMay lack co-factors that are important for protein function and/or complex assembly.
***P. pastoris*** Can perform the majority of post translational modificationsInexpensive Lack some post-translational modificationsMay lack co-factors that are important for protein function and/or complex assembly.
**Baculovirus expression in insect cells** Can perform all post translational modificationsBaculovirus is less cytotoxic than other viral strains Preparing the virus for the transduction is time consumingThe virus is not stable and cannot be stored for long periods of timeInsect cells may lack co-factors that are important for protein function and/or complex assembly.
**Bioreactors with mammalian systems** Can perform all post translational modificationsCan provide all co-factors needed for protein function and complex assemblyCan culture cells at very high densitiesCan express all mammalian proteins that require complex folding machineriesVery high protein yields Requires specialized staff to use the bioreactor.Expensive equipment is required.Optimizing expression conditions may be time consuming.
**HEK 293F suspension cells** Can perform all post translational modificationsCan provide all co-factors needed for protein function and complex assemblyCan express all mammalian proteins that require complex folding machineryFast and intuitive transfection protocolRelatively inexpensive compared to bioreactors and/or other commercially available transfection reagents. Although affordable, not as cheap as expression in bacteria
-------------------------------------------- ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
**Table 2. Advantages and disadvantages of main expression systems.**
Discussion
==========
We have developed a straightforward and cost effective method (PEI costs much less than commercially available lipophilic transfection reagents) for expressing and purifying large amounts of recombinant proteins and multi-subunit complexes from mammalian cells. Optimal transfection and expression efficiency can be reached if highly pure plasmid DNA (260/280 between 1.8 and 2.0) is used in combination with PEI as described in the protocol section. Cells must be cultured in serum- and antibiotic-free media, therefore, sterile technique is strictly required for passaging and transfecting cells in order to avoid costly and time-consuming infections. Cell viability should be 90% or higher and cultures should not be grown to densities greater than 2.5 x 10^6^cells/ml immediately before a transfection, as doing so will reduce protein yield. For a successful transfection, cultures should only contain single or dividing cells. Clusters may be broken up by vigorous vortexing 20 to 30 sec (**Figure 5**). The ratio of each plasmid used in the co-transfection can be varied by the user according to the stoichiometry of the complex being studied. Expression efficiency can be optimized for the protein of interest, so that a suitable expression time can be established. Purification should be performed keeping the protein sample cold at all times to reduce the risk of unwanted proteolytic degradation. The choice of tags and purification buffers is critical, since they may interfere with structural elements or active sites of certain proteins, often resulting in reduced solubility and/or loss of enzymatic activity. Small scale transfections are particularly useful for testing different constructs for the purification of large protein complexes.
Today, a large variety of alternative methodologies are available for the expression of recombinant eukaryotic proteins (**Table 2**). For example, Baculovirus expression in insect cells is widely used due to its high transduction efficiency and its lack of cytotoxicity in comparison to other viral species^6^. However, the process of making the virus is time-consuming and its instability does not allow the virus to be stored for long periods. On the other hand, yeast expression systems offer the possibility of growing cells to very high densities in fermenter cultures, resulting in high protein yields^7^. But they still lack the full variety of post-translational modifications required for eukaryotic proteins to fold correctly and interact with their protein partners. HDAC3, for example, is expressed but does not fold in *E. coli*. However, when expressed in 293F cells, we were able to purify an active enzyme in complex with its co-repressor (SMRT) and a molecule of inositol tetraphosphate (IP4)^8^, which is not found in prokaryotic cells. This method has also allowed us to purify and solve the crystal structure of HDAC1 interacting with MTA1 in the NuRD complex, which is similarly activated by IP4^9^. Ideally, we would always like to express eukaryotic proteins in their natural, physiological environment. Mammalian expression systems employing HEK 293-EBNA1 cells in bioreactors^10-12^ have been described and yield very high levels of protein, but these can be complex to use.
Our method is a simple and accessible alternative to expression in bacteria, yeast and bioreactor systems. The expression method is fast and does not require the use of expensive equipment, particularly if the roller bottle or flask atmosphere is replaced with 5% CO~2~and standard shaking incubators are used. We have used these protocols to co-transfect multiple plasmids and purify complexes with up to five proteins with yields of \>1 mg/L of culture. Interestingly, the system helps the identification of stable well-behaved complexes. For instance, expression of the binary complex of HDAC1 and Sin3A gave limited yields of protein, yet addition of SDS3 resulted in 5-fold higher amounts and hence guides the structural biologist in choice of suitable constructs and stable complexes for crystallization.
Disclosures
===========
The authors declare that they have no competing financial interests.
We would like to thank Dr. Xiaowen Yang (PROTEX cloning service) for preparing the expression constructs used for this work and the University of Leicester Core Biotechnology Services. This work was supported by a BBSRC studentship, the Wellcome Trust Programme and Senior Investigator grants WT085408 & WT100237 and BBSRC grant RM31G0224.
[^1]: Correspondence to: John W.R. Schwabe at <[email protected]>
|
{
"pile_set_name": "PubMed Central"
}
|
George W. Bush on Labor & Farming
George W. Bush on Farm Policy
Don’t use food as a diplomatic weapon; it hurts farmers
Q: The family farms are disappearing and having a hard time, even in the current positive economic environment. Your plan?
BUSH: I’d like our farmers feeding the world. We’re the best producers in the world. And I want the farmers feeding the world.
We need to open up markets. Exports are down. And every time an export number goes down, it hurts the farmer. I want the next president to have fast track negotiating authority to open up markets around the world. We’re the best. We’re the most efficient
farmers. I don’t want to use food as a diplomatic weapon from this point forward. It hurts the farmers. It’s not the right thing to do.
We need more work on value-added processing. I’m for research and development, so that we can use our technological
base to figure out new uses for farm products.
I’m for completely getting rid of the death tax. One reason family farmers are forced to sell early is because of the death tax. It’s a tax that taxes people twice, it penalizes the family farmer.
Source: St. Louis debate
Oct 17, 2000
Keep Freedom to Farm Act: cut price supports
Q: The Freedom to Farm Act cut price supports to farmers while giving them more freedom to plant what they want. Do you support this law?
A: Yes. The best way to ensure a strong and growing agricultural sector is through a
more market-driven approach that allows our farmers to fully participate in the world economy. As farming moves towards market-driven production, I believe the government should help farmers adapt to a global marketplace by providing
a strong safety net and the means to manage the cyclical downturns in the farm economy. I will reinforce the important role farmers and ranchers play in the U.S. economy by increasing trade opportunities, reducing regulatory
burdens, and reducing the overall tax burden. My administration will also renew our commitment to investments in new and innovative technologies for rural America.
Source: Associated Press
Sep 20, 2000
Farm policy: emergency relief; insurance; antitrust
From this point forward, would not use food as a unilateral sanction or diplomatic weapon
Called for emergency disaster relief, both through direct payments to farmers and through reforming crop insurance
Supports tax incentives for use
of ethanol
Will ensure aggressive enforcement of antitrust laws and will move forcefully against any agribusiness entities that are conducting anti-competitive practices
Source: GeorgeWBush.com: ‘Issues: Policy Points Overview’
Apr 2, 2000
Family farms: Food for Peace instead of food as a weapon
Q: Since the family farmer is self-employed, would you cap government agriculture benefits to a modest one-family level? A: I would look at the formulas to make sure that the money was distributed fairly. And I would have an agricultural department that
would send the money out on a timely basis. I believe we ought to open up markets all around the world. We ought to reduce barriers and tariffs. We shouldn’t be using food as a diplomatic weapon. We ought to implement the food for peace program.
We ought to eliminate the death tax as well so people can pass their farm from one generation to the next. And we ought to have good sound risk management policies that give farmers more options when it comes to crop insurance, and more options on how to
manage their income. Agriculture is incredibly important for this country and one of the reasons why we’ve had trouble in the world is because administrations have traded off agriculture just as if it’s a secondary part of our economy. It’s not.
Source: GOP Debate in Johnston, Iowa
Jan 16, 2000
Ethanol tax incentives good for farmers and good for air
George W. Bush says that he “supports tax incentives for use of ethanol [because] not only is it good for the farmer, it is good for the quality of air all across America.”
George W. Bush on Labor
Flex-time & family-oriented work rules via tax code
Bush put a family-oriented twist on a familiar theme, casting his $1.3 trillion, 10-year tax-cut proposal as part of a grander concept of providing greater flexibility to employees. He added several points to his agenda:
letting employees receive
compensation time instead of overtime pay;
making employer-purchased home computers tax deductible;
forbidding the Occupational Safety and Health Administration from monitoring home offices;
and encouraging television networks to return to an
evening “family hour.”
The overtime proposal, sometimes called “flex-time” by the Clinton administration, has been criticized by some Democrats who fear employers would force employees to take time off instead of overtime pay. A Bush aide said the
proposal would forbid such pressuring, but he did not elaborate on how the law would work. The House has passed legislation similar to Bush’s proposal twice, and Clinton has supported a narrower version. Unions are largely opposed to any measure.
|
{
"pile_set_name": "Pile-CC"
}
|
A response to a query submitted under the Right to Information (RTI) Act has, reportedly, revealed that the Lokpal selection committee met for the first time in March 2018, nearly four years after the Bharatiya Janata Party (BJP) came to power in 2014.
However, the RTI response declined to give the minutes of the six meetings that have been held since March, under the chairmanship of Prime Minister Narendra Modi. According to the government, the minutes cannot be made public since “high-level dignitaries” attended the meetings, which necessitated classifying the minutes as “secret documents”.
In its RTI response, the Department of Personnel and Training (DoPT) said, “As regards the minutes of the meetings it is submitted that the authorship of such documents which include 3-5 high level dignitaries does not vest in the Department of Personnel & Training and (the) same have been shared as secret documents. Thus copies of the said documents cannot be provided by the undersigned CPIO (Central Public Information Officer).”
This reason cited by the DoPT, however, appears to be a baseless one since the Public Information Officer (PIO) has not stated any exemptions under the RTI Act to decline providing the details asked in the RTI application. As such, this display of lack of credulousness adds to the imagery of the BJP government not being inclined towards appointing the Lokpal as it had promised in its 2014 election manifesto.
Also Read: Maintain Transparency in Appointment of CIC, ICs: SC to Centre, States
|
{
"pile_set_name": "OpenWebText2"
}
|
[Ultrasound controlled fine needle cytology and fine needle histology in circumscribed pancreatic processes].
Using a biopsy transducer 171 fine-needle punctures of focal pancreatic processes were done in 158 patients. The final diagnosis was made in the first group by surgical operation or autopsy, in the second by the further clinical course. The sensitivity of the cytological examination alone was 78% in the first group and 90% in the second. The histological examination alone had a sensitivity of 79% to 100%. Combining both ways of examination according to defined criteria the sensitivity in the first group was 84% and in the second 100%. The specificity was 100% for both. The complications were 2.9%. Of these 5 cases one pancreatitis had to be operated. There were no lethal complications. To optimize the sensitivity and specificity of the ultrasound guided fine needle puncture technique we want to recommend the following points: A biopsy transducer or puncture-aid must be available. The fine needle technique only should be done under the prerequisite of a high frequency of punctures and a high constancy in the relation between puncturer and pathologist. To receive representative material it can be necessary to make several needle passes during one session. Material for cytological und histological examination should always be taken simultaneously.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Alterations in the noradrenergic projection to the cerebellum of the dystonic (dt) rat.
The genetically dystonic rat (dt) has elevated resting levels of cerebellar norepinephrine (NE) in comparison with phenotypically normal littermates. This difference is not secondary to cerebellar hypoplasia. Increased NE is observed as early as postnatal day 12, when clinical symptoms have become evident. The elevation in cerebellar NE levels in the dt rat involves all cerebellar areas, but is not generalized to all terminal fields of the locus coeruleus. Elevations in cerebellar NE are followed developmentally by a reduction in sensitivity to the NE-depleting effects of reserpine, a change which is also confined to the cerebellum. The effects of amphetamine and the tyrosine hydroxylase inhibitor alpha-methyl-para-tyrosine were similar in normal and dt rats. Levels of the major cerebellar metabolite of NE, 3-methoxy-4-hydroxyphenylglycol, did not differ between mutant and normal animals. Nor were any changes noted in the number or affinity of beta-adrenergic receptors. These data indicate that there is a regional alteration in NE storage. Cerebellar morphology appears normal in the dt rat, except for a decrease in Purkinje cell size. This change and other evidence of biochemical abnormalities in the Purkinje cells suggest that the alterations in cerebellar NE in the dt mutant may be a secondary response to a functional change in the target neuron for this system, the Purkinje cell.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Modeling deformation behavior of the baseball.
Regulating ball response to impact is one way to control ball exit velocity in baseball. This is necessary to reduce injuries to defensive players and maintain the balance between offense and defense in the game. This paper presents a model for baseball velocity-dependent behavior. Force-displacement data were obtained using quasi-static compression tests to 50% of ball diameter (n = 70 baseballs). The force-displacement curves for a very stiff baseball (Model B) and a softer type (Model C) were characterized by a Mooney-Rivlin model using implicit finite element analysis (ANSYS software, version 6.1). Agreement between experimental and numerical results was excellent for both Model B (C(10) = 0, C(01) = 3.7e(6) Pa) and Model C (C(10) = 0, C(01) = 2.6e(6) Pa). However, this material model was not available in the ANSYS/LSDYNA explicit dynamic software (version 6.1) used to quantify the transient behavior of the ball. Therefore the modeling process was begun again using a linear viscoelastic material. G(infinity), the long-term shear modulus of the material, was determined by the same implicit FEA procedure. Explicit FEA was used to quantify the time-dependent response of each ball in terms of instantaneous shear modulus (G0) and a decay term (beta). The results were evaluated with respect to published experimental data for the ball coefficient of restitution at five velocities (13.4-40.2 ms(-1)) and were in agreement with the experimental values. The model forms the basis for future research on baseball response to impact with the bat.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Extract of the seed coat of Tamarindus indica inhibits nitric oxide production by murine macrophages in vitro and in vivo.
The seed coat extract of Tamarindus indica, a polyphenolic flavonoid, has been shown to have antioxidant properties. The present studies investigated the inhibitory effect of the seed coat extract of T. indica on nitric oxide production in vitro using a murine macrophage-like cell line, RAW 264.7, and in vitro and in vivo using freshly isolated B6C3F1 mouse peritoneal macrophages. In vitro exposure of RAW 264.7 cells or peritoneal macrophages to 0.2-200 microg/mL of T. indica extract significantly attenuated (as much as 68%) nitric oxide production induced by lipopolysaccharide (LPS) and interferon gamma (IFN-gamma) in a concentration-dependent manner. In vivo administration of T. indica extract (100-500 mg/kg) to B6C3F1 mice dose-dependently suppressed TPA, LPS and/or IFN-gamma induced production of nitric oxide in isolated mouse peritoneal macrophages in the absence of any effect on body weight. Exposure to T. indica extract had no effect on cell viability as assessed by the MTT assay. In B6C3F1 mice, preliminary safety studies demonstrated a decrease in body weight at only the highest dose tested (1000 mg/kg) without alterations in hematology, serum chemistry or selected organ weights or effects on NK cell activity. A significant decrease in body weight was observed in BALB/c mice exposed to concentrations of extract of 250 mg/kg or higher. Oral exposure of BALB/c mice to T. indica extract did not modulate the development of T cell-mediated sensitization to DNFB or HCA as measured by the local lymph node assay, or dermal irritation to nonanoic acid or DNFB. These studies suggest that in mice, T. indica extract at concentrations up to 500 mg/kg may modulate nitric oxide production in the absence of overt acute toxicity.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
ACLU: Squalid conditions at troubled Baltimore jail
BALTIMORE (AP) — Moldy showers and cells are infested with mice and cockroaches, toilets are broken for days and medical care is poor or nonexistent at a notoriously troubled Baltimore city jail, the ACLU and Public Justice Center said in court documents filed Tuesday.
The groups want a federal judge to reopen a lawsuit against the state of Maryland over what they say are conditions so substandard it brings "shame to this city." Inmates suffering from serious illnesses such as HIV and diabetes are being denied life-sustaining prescription medication and the state has failed to cure well-documented systemic problems within the jail after entering into a 2007 agreement with the U.S. Department of Justice, the court documents said.
"The conditions at the jail are a menace to the Constitution and to the public's health," said David Fathi, director of the ACLU's National Prison Project.
Stephen Moyer, secretary of the Maryland Department of Public Safety and Correctional Services, said he was committed to making changes. The department noted the state has spent more than $58 million to improve the safety and security of inmates and staff over the past 10 years.
"I was 4 years old when this original case was filed in 1964," Moyer said in a statement. "I am 54 and have only had four months to learn about these perceived issues. I have high expectations of my staff to change the way we do business."
The Baltimore City Detention Center, known as BCDC, grabbed headlines in 2013 after a sweeping federal indictment exposed a sophisticated drug and cellphone smuggling ring inside the jail involving dozens of gang members and corrections officers, four of whom were impregnated by the gang member who orchestrated the scheme.
The motion argues that widespread failures of the jail's medical and mental health care have possibly played a role in the death of seven inmates over the last couple of years.
The 103-page motion, which asks the court to set a hearing date, cites multiple cases in which inmates failed to get medical treatment. In one example, a man died after experiencing a three-day delay in renewal of his high blood pressure medication.
The jail "remains a dank and dangerous place, where detainees are confined in dirty cells infested with vermin. The showers are full of drain flies, black mold and filth," the motion said. The facility holds about 3,000 people on a given day, and the oldest part of the building dates back to 1859.
Inmates with disabilities are assigned to cells that can't meet their needs and they are often denied health care supplies, ranging from urinary bag and catheter changes to properly working wheelchairs, the documents said.
There were 43 incidents involving detainees making urgent medical visits related to excessive heat because their cells were not properly ventilated, reaching a heat index of 94 degrees in one instance, according to the motion. Over a three-month period, there were 26 instances in which broken toilets were not repaired for three days, far longer than the eight-hour mandate.
"We're asking the court to reopen the case so we can litigate in court the defendants' failure to comply with the settlement agreement and failure to provide what's minimally required under the Constitution, in terms of medical care and safety," said Debra Gardner, legal director for the Public Justice Center.
|
{
"pile_set_name": "Pile-CC"
}
|
By Tim Durham
Dr. Durham’s family operates a 30-acre vegetable farm on Long Island, New York — where he returns to work every summer. During the academic year, Dr. Durham teaches general biology, plant science, biotechnology, genetics, and agronomy at Ferrum College in Virginia.
Science is a framework of discovery and application to tackle real-world problems. Since the discovery of DNA, our knowledge has increased exponentially! But how do we put that to work? Biotechnology is applied molecular genetics. Scientists observed a number of phenomena “in the wild” and modeled these methods for crop improvement.
For example, scientists observed a bacterium called Agrobacterium tumefaciens genetically engineering its host plants. It causes a disease called Crown Gall (stem growths). Basically, it adds some extra directions (in the form of genes) to the plant’s genetic blueprint. These genes instruct the plant to produce food for the bacterium! The galls are the observable result of this spot genetic engineering — miniature (rewritten) cellular factories.
Intrigued, scientists then piggybacked this discovery with another, restriction enzymes in bacteria. Restriction enzymes are used by bacteria to cut the DNA of invading viruses. This defensive approach basically slices and dices them into genetic bits, effectively inactivating them. What scientists found was that different enzymes cut in very specific ways.
So let’s say organism X has genes of interest that could improve a crop. These genes can be duplicated and cut out with a specific restriction enzyme. In parallel, the undesirable genes that cause crown gall in Agrobacterium are cut out. We can then paste the genes from organism X, filling the gap we just created in Agrobacterium. Rejoin the cut ends and voila, the Agrobacterium is now a vector for the genes we want it to deliver into a plant — taking a page out of the Windows PC “cut and paste” function.
There are other methods to produce GMOs, including the “gene gun”, but Agrobacterium (nature’s genetic engineer) was the initial inspiration. A case study in modeling nature — biomimicry!
Whatever the method, GMOs greatly streamline farm operations, including herbicide tolerance (Roundup Ready technology) or bt traits that equip the plant with its own biodegradable insecticide. One nibble and pest caterpillars or rootworms are toast. Many more applications are in the works, including nutrient fortification — which admittedly has more public appeal.
In theory, sound science should counterbalance controversy. In practice, this has not been the case with GMOs. Despite a decades-long track record of safety, they’re still treated harshly by regulators. The spiritual sequel to GMOs are in the works — gene editing. Rather than cut and paste outside genetic materials into an organism, as with GMOs, gene editing simply rewrites what’s already there. This is poised to avoid much of the current controversy.
|
{
"pile_set_name": "OpenWebText2"
}
|
UN official calls for reversal of ‘Allah’ ban
Freedom of religion or belief is a right of human beings, not a right of the state,says UN special rapporteur on freedom of religion or belief Heiner Bielefeldt.
KUALA LUMPUR: Malaysia should reverse a ban on a Christian newspaper using the word ‘Allah’ to refer to God, a UN official said on Monday about a decision that fanned religious tension in the country.
Malaysia’s second highest court ruled in October that the Catholic newspaper, the Herald, could not use the word ‘Allah’ to refer to God, in a landmark decision on an issue that has fanned religious tension in the majority Muslim country.
“Freedom of religion or belief is a right of human beings, not a right of the state,” the UN special rapporteur on freedom of religion or belief, Heiner Bielefeldt, said in statement.
“It cannot be the business of the state to shape or reshape religious traditions, nor can the state claim any binding authority in the interpretation of religious sources or in the definition of the tenets of faith,” he said.
The chief judge in the Malaysian court said the word Allah was not an integral part of the faith in Christianity and that its use by the newspaper would cause confusion.
Since then, confusion has reigned over the interpretation of the ruling, with government ministers, lawyers and Muslim authorities giving widely diverging views on its scope.
Critics of the ruling have said it casts a chill on religious and minority rights in Malaysia, which has substantial minorities of non-Muslim ethnic Chinese and Indians.
“The current case may affect the right of all non-Muslims in Malaysia to use the word ‘Allah’ while referring to God,” the UN official said.
Lawyers for the Catholic paper had argued that the word Allah predated Islam and has been used for centuries by Malay-speaking Christians in Malaysia’s part of Borneo island.
They said they planned to appeal against the decision in the country’s highest court.
Christians make up about 9 percent of Malaysia’s 28 million people. Muslims are the majority at about 60 percent.
|
{
"pile_set_name": "Pile-CC"
}
|
The invention concerns the automatic weighing method of laundry in washing machine by specific measurements performed during different angular speeds of its drum.
|
{
"pile_set_name": "USPTO Backgrounds"
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.