id
int64 0
12.9M
| type
large_stringclasses 5
values | by
large_stringlengths 2
15
⌀ | time
timestamp[us] | title
large_stringlengths 0
198
⌀ | text
large_stringlengths 0
99.1k
⌀ | url
large_stringlengths 0
6.6k
⌀ | score
int64 -1
5.77k
⌀ | parent
int64 1
30.4M
⌀ | top_level_parent
int64 0
30.4M
| descendants
int64 -1
2.53k
⌀ | kids
large list | deleted
bool 1
class | dead
bool 1
class |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
41,813,100 | comment | nisten | 2024-10-11T20:01:19 | null | Is hackernews..hacked? | null | null | 41,812,523 | 41,812,523 | null | null | null | null |
41,813,101 | comment | bumby | 2024-10-11T20:01:23 | null | ><i>The problem is that general problem solving requires potentially arbitrary amounts of moving data</i><p>Can you expand on this thought? | null | null | 41,812,897 | 41,812,523 | null | null | null | null |
41,813,102 | comment | rory | 2024-10-11T20:01:35 | null | By chance I just talked to someone with deeper knowledge on this and they said the current constraint is actually ramping up supply of the delivery mechanism, not the drug.<p>I have zero expertise on this, but would be curious if anyone knows what's special about Ozempic delivery that can't be served by a commodity syringe. | null | null | 41,811,936 | 41,811,263 | null | [
41813383
] | null | null |
41,813,103 | comment | JSDevOps | 2024-10-11T20:01:36 | null | This was such a fun tool back in the day. | null | null | 41,812,003 | 41,812,003 | null | null | null | null |
41,813,104 | comment | apwell23 | 2024-10-11T20:01:36 | null | Have you noticed any effect on gastric emptying. As someone with 'tummy issues' ( ibs/gerd ect) i am apprehensive of messing with my digestive systems. | null | null | 41,812,493 | 41,811,263 | null | null | null | null |
41,813,105 | story | null | 2024-10-11T20:01:38 | null | null | null | null | null | 41,813,105 | null | null | true | true |
41,813,106 | comment | glomgril | 2024-10-11T20:02:01 | null | He is coming from the perspective of a long-running debate on symbolic versus statistical/data-driven approaches to modeling language structure and use. It seems in recent years he has had trouble coming to terms with the fact that at least for real-world applications of language technology, the statistical approach has simply won the war (or at worst, forms the core foundation on top of which symbolic approaches can have some utility).<p>I come from the same academic tradition, and have colleagues in common with him. He has been advocating for a quasi-chomskyan perspective on language science for many years -- as have many others working at the intersection of linguistics and psychology/cog sci.<p>TBH I suspect he himself is a large part of his target audience. A lot of older school academics raised in the symbolic tradition are pretty unsettled by the incredible achievements of the data-driven approach.<p>Personally I saw the writing on the wall years ago and have transitioned to working in statistical NLP (or "AI" I suppose). Feeling pretty good about that decision these days.<p>FWIW I do think symbolic approaches will start to shine in the next several years, as a way to control the behavior of modern statistical LMs. But doubtful they will ever produce anything comparable to current systems without a strong base model trained on troves of data.<p>edit: Worth noting that Marcus has produced plenty of high-quality research in his career. I think his main problem here is that he seems to believe that AI systems should function analogously to how human language/cognition functions. But from an engineering/product perspective, <i>how</i> a system works is just not that important compared to <i>how well</i> it works. There's probably a performance ceiling for purely statistical models, and it seems likely that some form of symbolic machinery can raise that ceiling a bit. Techniques that <i>work</i> will eventually make their way into products, no matter which intellectual tradition they come from. But framing things in this way is just not his style. | null | null | 41,812,932 | 41,812,523 | null | null | null | null |
41,813,107 | comment | tcsenpai | 2024-10-11T20:02:01 | null | Personally I use llama3.1:8b or mistral-nemo:latest which have a decent contex window (even if it is less than the commercial ones usually). I am working on a token calculator / division of the content method too but is very early | null | null | 41,812,624 | 41,810,507 | null | null | null | null |
41,813,108 | comment | neonsunset | 2024-10-11T20:02:05 | null | Noting that the article's findings from 2018 need to be re-evaluated on up-to-date versions before deriving conclusions because in the last 6 years (and especially in the last 3 or so for .NET) the garbage collector implementations of both Go and .NET have evolved quite significantly. The sustained multi-core allocation throughput graph more or less holds but other numbers will differ significantly.<p>One of the major factors that play in Go's favour is the right attitude to architecting the libraries - the zero-copy slicing is much more at the forefront in Go than in .NET (technically incorrect but not in terms of how the average impl. looks like), and the flexible nature of C# combined with it being seen as "be glad we even support this Microsoft's Java" by many vendors lead to poor quality vendor libraries. This results in the experience where developers see Go applications be more efficient, not realizing that it's the massively worse quality implementation of a dependency their .NET solution has to deal with (there was a recent comparison video, where .NET was estimated to be slower, but the reality was that it wasn't .NET but the AWS SDK dependency and the benchmark author being most familiar with Go and making optimal choices with significant impact there like using DB connection pooling).<p>I'm often impressed by how much punishment GC and compiler can take, continuing to provide competitive performance despite massive amounts of data reallocations and abstraction bloat thrown at it by developers who don't want to even consider to approach C# in an idiomatic C# way (at the very least by listening to IDE suggestions and warnings). In some areas, I even recommend to look at community libraries first which are likely to provide far superior experience if documentation and brief code audit indicate that its authors care(tm) which is one of the most important metrics. | null | null | 41,812,387 | 41,787,041 | null | null | null | null |
41,813,109 | comment | sebastialonso | 2024-10-11T20:02:05 | null | I don't think that's the point of the article, happy to read the reasoning used to get to that conclusion.<p>To me the point is not concerned with usefulness, is with reliability. You could get correct answers out of the agent, but how often do you get correct data versus gibberish? It's an extremely important metric to consider, and it's the same reason you wouldn't hop into a self-driving car in the real world if it can drive flawlessly in a straight line, but once every three intersections turns the wrong way. | null | null | 41,812,863 | 41,812,523 | null | null | null | null |
41,813,110 | comment | ChrisArchitect | 2024-10-11T20:02:33 | null | Related:<p><i>Next.js, Just Why? (2023)</i><p><a href="https://news.ycombinator.com/item?id=41803327">https://news.ycombinator.com/item?id=41803327</a> | null | null | 41,812,842 | 41,812,842 | null | null | null | null |
41,813,111 | comment | rfrey | 2024-10-11T20:02:39 | null | Of course I eat carbs when I shouldn't. Not the bread, but I eat potatoes sometimes, and too much fruit. I'm not denying that I eat too much.<p>The point is you claim that if we gluttons would just cut out sugar for 2 weeks and learn to be an adult, our appetites and cravings would disappear. That's nonsense, and your dismissal of data that doesn't fit your narrative makes your accusations towards others of being anti-science both hollow and ironic. | null | null | 41,812,193 | 41,811,263 | null | null | null | null |
41,813,112 | comment | AdamJacobMuller | 2024-10-11T20:02:40 | null | Google made something better than what existed with Chrome, it was obvious it would capture the market significantly especially among more technical people.<p>I don't think the fact that Chrome is (was) better is the question, it's a question of how they got here.<p>Google took tons of money and threw it into Chrome and therefore developed something better. It's better because Google put more money into it than anyone else would have because, in the absence of considering using it to enshrine their search and ad revenue, it wouldn't make sense.<p>Isn't this part of the antitrust test? | null | null | 41,813,003 | 41,809,698 | null | [
41813354
] | null | null |
41,813,113 | comment | tzs | 2024-10-11T20:02:42 | null | Pamphlets cost money to print and distribute, and distribution channels were not fast. Most crazy people didn't have the money for widespread distribution, and the slowness of distribution meant that if some insane thing did start getting traction others could get debunking material out before the crazy thing got too far.<p>With social media the distribution cost is zero to the person posting, so there is no speed bump there. It is distributed in a very short time to a large number of people, so there is little opportunity to debunk it before it has taken hold.<p>The only real speed bump with social media is the time to write the material. That's generally much lower for the insane stuff then for the correct stuff.<p>Try debunking sometime if you want to see this yourself. It can take hours or days of work to write a good refutation for something that a crazy person spent a few minutes on. And then when you finish, you often find the crazy person has made a dozen more arguments and their audience dismisses your debunking saying everyone occasionally makes a mistake and pointing at the 10 new arguments that have not been debunked yet.<p>With pamphlets the aforementioned limitations also meant that you generally didn't see the pamphlets from crazy people far away. You mostly just saw your local or regional sources. Overall the information you received was mostly sane stuff with an insane pamphlet now and then.<p>With social media you get the insanity from worldwide. You might find that a large majority of your feed is the crazy stuff so even if someone writes a good refutation that would have convinced you it can easily be overlooked in the deluge of world wide craziness. | null | null | 41,810,362 | 41,807,121 | null | [
41813335
] | null | null |
41,813,114 | story | mpweiher | 2024-10-11T20:02:42 | Writing the Worst Datalog Ever in 26loc | null | https://buttondown.com/tensegritics-curiosities/archive/writing-the-worst-datalog-ever-in-26loc/ | 1 | null | 41,813,114 | 0 | null | null | null |
41,813,115 | comment | willcipriano | 2024-10-11T20:02:44 | null | Damn, 2 grams? I had no idea the dose could get that high. | null | null | 41,812,997 | 41,812,876 | null | null | null | null |
41,813,116 | comment | apwell23 | 2024-10-11T20:02:46 | null | They don't need marketing team writing posts on HN. The drug sells itself and even shortages all over the world. | null | null | 41,813,078 | 41,811,263 | null | [
41813368
] | null | null |
41,813,117 | comment | jbverschoor | 2024-10-11T20:02:48 | null | Well, the web is unusable without an adblocker.
Time to move to another browser. | null | null | 41,809,698 | 41,809,698 | null | null | null | null |
41,813,118 | comment | lazyeye | 2024-10-11T20:02:54 | null | Thanks for all your efforts.
Will definitely check this out. | null | null | 41,801,323 | 41,794,577 | null | null | null | null |
41,813,119 | story | jasondavies | 2024-10-11T20:02:57 | Diffusion for World Modeling: Visual Details Matter in Atari | null | https://diamond-wm.github.io/ | 1 | null | 41,813,119 | 0 | null | null | null |
41,813,120 | comment | toomuchtodo | 2024-10-11T20:02:59 | null | To add to this, adult relationships are shades of shared adversity, reciprocity, vulnerability, and demonstrating reliability. Avoid transactional behavior, put yourself out there, accept losses but be present in the positive moments. | null | null | 41,812,944 | 41,810,889 | null | null | null | null |
41,813,121 | comment | AlienRobot | 2024-10-11T20:03:02 | null | What is moving data round? Isn't everything in a computer moving data around? Do you mean backpropagation or somtehing more specific? | null | null | 41,812,897 | 41,812,523 | null | null | null | null |
41,813,122 | story | idrt | 2024-10-11T20:03:20 | 11111111111111111111111111111111 | null | https://www.web.com/blog/id | 1 | null | 41,813,122 | 0 | null | null | null |
41,813,123 | comment | zb3 | 2024-10-11T20:03:22 | null | Manifest V3 is not the problem itself. But removing webRequestBlocking and creating some ridiculous limits for the DNR api are, these changes should be reversed downstream. | null | null | 41,809,698 | 41,809,698 | null | null | null | null |
41,813,124 | comment | sockaddr | 2024-10-11T20:03:22 | null | No shit, Gary.<p>Every time I see this guy pop up is some bad take or argument with someone. What’s the deal with him? | null | null | 41,812,523 | 41,812,523 | null | null | null | null |
41,813,125 | comment | camgunz | 2024-10-11T20:03:25 | null | 80% of everything is crap, including psychiatrists, plus if you take away SSRIs that's a huge chunk of their business. Upton Sinclair and all that. | null | null | 41,813,083 | 41,812,876 | null | null | null | null |
41,813,126 | comment | Scoundreller | 2024-10-11T20:03:28 | null | Being full of food also makes it hard to exercise. | null | null | 41,812,372 | 41,811,263 | null | null | null | null |
41,813,127 | comment | maicro | 2024-10-11T20:03:33 | null | Hmm, I'm not saying it's a good idea, but what about a daemon that keeps a symlinked version of the entire jira environment up to date? So you have one jira-as-filesystem that's the raw files, but then for human consumption/interaction, you have a tree of symlinks, including multiple links to the same file wherever it's relevant. Might be adding more layers than needed, based on my lack of understanding, but might technically solve the (current/stated) abstraction issue. | null | null | 41,812,623 | 41,811,983 | null | [
41813206,
41813353
] | null | null |
41,813,128 | comment | bonzini | 2024-10-11T20:03:37 | null | Because they had enough of Trump after 4 years.<p>Dick Cheney is voting for Harris, that's the equivalent of Norman Osborn voting for Peter Parker. | null | null | 41,811,387 | 41,807,681 | null | null | null | null |
41,813,129 | comment | fireflash38 | 2024-10-11T20:03:44 | null | For some people it is clinical. For others it is mental/willpower. That said... It's exceptionally difficult in the modern world to do everything necessary to be at a healthy weight. Things are shoved at you constantly that are terrible for you. It's so, so much easier to eat poorly and to excess. Combine that with dopamine hits from consuming sugar/fat? No surprise people overeat. | null | null | 41,812,839 | 41,811,263 | null | null | null | null |
41,813,130 | comment | anigbrowl | 2024-10-11T20:03:48 | null | Just eat regular food. Hardly anything in my diet would be mysterious to a time traveler from 100 years ago, or indeed 200 years ago. Occasionally I enjoy chocolate or instant noodles but mostly it's just fruit, veg, meat, dairy, and grains. If you don't know what something is made out of or how, then don't eat it. | null | null | 41,812,873 | 41,811,263 | null | null | null | null |
41,813,131 | comment | metadaemon | 2024-10-11T20:03:48 | null | Same with SSNIs, I get brain zaps if I don't take my medication early enough in the day. It can come with nausea too, so missing a dose can potentially ruin my evening. | null | null | 41,812,876 | 41,812,876 | null | null | null | null |
41,813,132 | story | chrchr | 2024-10-11T20:03:48 | U.S. Web Design System: Introducing Uswds 3.0 (2022) | null | https://designsystem.digital.gov/whats-new/updates/2022/04/28/introducing-uswds-3-0/ | 1 | null | 41,813,132 | 0 | [
41813141
] | null | null |
41,813,133 | comment | detourdog | 2024-10-11T20:03:50 | null | I’m happy to defend them. I just don’t know what you think is correct. I understand you believe I’m wrong. I see the linkage as crystal clear evolution of human thought.<p>You have some other ideal I assume.<p>I see you as unwilling to defend your ideas. | null | null | 41,813,077 | 41,776,631 | null | null | null | null |
41,813,134 | comment | itronitron | 2024-10-11T20:03:51 | null | This entire thread is a marketing campaign for Ozempic and other drugs. If you read the comments critically they read just like ad copy. | null | null | 41,812,939 | 41,811,263 | null | [
41813207,
41813192
] | null | null |
41,813,135 | comment | null | 2024-10-11T20:03:51 | null | null | null | null | 41,803,327 | 41,803,327 | null | null | true | null |
41,813,136 | comment | hilux | 2024-10-11T20:03:54 | null | Living a healthy lifestyle, so that you won't need Ozempic, is possible - even surrounded by all the drugs and booze and carbs.<p>I do it. Lots of people do it. The information is freely available, now more than ever.<p>From all I've read, Ozempic isn't a silver bullet, and has many side effects. It just helps people stave off the inevitable for a while. | null | null | 41,811,263 | 41,811,263 | null | null | null | null |
41,813,137 | comment | stevenAthompson | 2024-10-11T20:04:12 | null | >That's how all creatures in the wild do it... Wait, no, it isn't.<p>Neither creatures in the wild, or primitive man, have access to unlimited quantities of calorie dense foods. We could go back to that lifestyle, but billions would have to die and the overall human lifespan would decrease rather than increasing.<p>I think I'd rather take a perfectly safe drug than go back to wiping with leaves and hunting for worm riddled meat.<p>> How much does that lifestyle sound like freedom to you?<p>I'm not even sure I understand which lifestyle you're asking about here, but if you mean the modern lifestyle then it's certainly more free than the lives primitive man had. "Might makes right" was the rule of the land back then, and contrary to your imagination, you probably wouldn't have been the mightiest. Certainly not forever.<p>Hell, it's more free now than the lives most of our grandparents had. 50 years ago about half of all white people surveyed said they'd move away if a black person bought a house in their neighborhood, and gay people were routinely murdered for existing.<p>There were no "good old days", and Stardew Valley is just a game. | null | null | 41,812,929 | 41,811,263 | null | null | null | null |
41,813,138 | comment | wongarsu | 2024-10-11T20:04:19 | null | In formal reasoning it's entirely valid to refute a hypothesis without providing a valid alternative. | null | null | 41,812,992 | 41,812,523 | null | null | null | null |
41,813,139 | story | mfiguiere | 2024-10-11T20:04:29 | Steam now says the 'game' you're buying is just a license | null | https://www.theverge.com/2024/10/11/24267864/steam-buy-purchase-license-digital-storefront | 2 | null | 41,813,139 | 2 | [
41813293,
41813297
] | null | null |
41,813,140 | comment | alaithea | 2024-10-11T20:04:31 | null | Would love to read this, but the site is down.<p>I weaned myself off of Sertraline and pursued OTC options just because the brain zaps and the yawns/drowsiness were so bad. Didn't matter what time of day I took it, didn't matter that the dosage was low. The brain zaps made me lose trust in my own faculties. These momentary, split-second losses of consciousness, where after each one, I'd have to spend another split-second reorienting myself to the environment, got way too disorienting. It also got worse the longer I was on it.<p>Finally weaned myself off and use SAM-e instead. No perceivable side-effects there. For anyone who doesn't know, SAM-e is an OTC supplement in the U.S. but the same chemical compound as one of the front-line antidepressants in Europe. | null | null | 41,812,876 | 41,812,876 | null | [
41813225,
41813154
] | null | null |
41,813,141 | comment | null | 2024-10-11T20:04:51 | null | null | null | null | 41,813,132 | 41,813,132 | null | null | true | null |
41,813,142 | comment | lcnPylGDnU4H9OF | 2024-10-11T20:05:06 | null | > In fact, it’s abundantly clear we won’t be satisfied until we’ve completely destroyed human intelligence as superior.<p>One could argue this is precisely where the goal posts have been for a long time. When did the term "singularity" start being used in the context of human technological advancements? | null | null | 41,813,066 | 41,812,523 | null | null | null | null |
41,813,143 | comment | pama | 2024-10-11T20:05:14 | null | I’m certain you’re joking here, but I wanted to add that multiplying a few digits is learned naturally from data without any trouble. Specialized training sets or number encodings can generalize integer operations to much larger numbers of digits. However, an infinite number of digits is not possible. Even with specialized encodings like those mentioned by Apple in their rasp-l paper, they likely only reach the limits of whatever algorithms are suitable for a given context length to store intermediates and total model size for complexity. | null | null | 41,812,992 | 41,812,523 | null | null | null | null |
41,813,144 | comment | thescriptkiddie | 2024-10-11T20:05:23 | null | I've gotten them from taking SNRIs, so I wonder if norepinephrine is involved somehow, or maybe SNRIs are not actually that selective. | null | null | 41,813,083 | 41,812,876 | null | [
41813174,
41813162
] | null | null |
41,813,145 | story | PaulHoule | 2024-10-11T20:05:57 | Artificial cell nuclei created in living egg from purified DNA injection | null | https://phys.org/news/2024-10-artificial-cell-nuclei-egg-purified.html | 2 | null | 41,813,145 | 0 | null | null | null |
41,813,146 | comment | amarshall | 2024-10-11T20:06:21 | null | Well the article brings up containers as an example. If the sysadmin controls “your” parent or root process (e.g. the login shell), they can just perform seccomp filtering there and it applies to everything within it (like any other sandbox). | null | null | 41,812,459 | 41,788,426 | null | null | null | null |
41,813,147 | comment | pixl97 | 2024-10-11T20:06:22 | null | >Large epidemiological studies have shown the link between obesity and pancreatic cancer. A large population-based case-control study of pancreatic cancer demonstrated that obesity was associated with a statistically significant 50–60% increased risk of pancreatic cancer.<p>So does obesity, so there is that. | null | null | 41,811,571 | 41,811,263 | null | null | null | null |
41,813,148 | comment | pyre | 2024-10-11T20:06:25 | null | Sounds like the real issue is "we are replacing X with Y" and there are use-cases for both X and Y to co-exist. | null | null | 41,812,852 | 41,809,698 | null | null | null | null |
41,813,149 | comment | timeon | 2024-10-11T20:06:26 | null | Oh I wish. | null | null | 41,810,536 | 41,809,698 | null | null | null | null |
41,813,150 | comment | ben_w | 2024-10-11T20:06:27 | null | They're already operating on an architecture that can do that for about a nanojoule.<p>You can also just ask them to write code for you, which appears to be what ChatGPT does now — it has its own python environment, I'm not sure what's in it except matplotlib and pandas, but it's at least that. | null | null | 41,812,992 | 41,812,523 | null | null | null | null |
41,813,151 | comment | yieldcrv | 2024-10-11T20:06:28 | null | This is interesting as this is my default, and I am pretty lean. I often have to remember to eat or check if I did, and lament that its a chore. I enjoy food and the dining experience, but I would be fine with that being an entertainment option once a week too.<p>I wonder if there is enough research on how gut bacteria influences these things, because if this is what people want maybe I could sell mine. | null | null | 41,812,603 | 41,811,263 | null | null | null | null |
41,813,152 | comment | ryukafalz | 2024-10-11T20:06:33 | null | Right. Design our cities so that getting around while getting exercise is safe and easy. And restaurants/etc should give you healthy portion sizes by default, not rely on your self-control to stop eating. And so on... | null | null | 41,812,886 | 41,811,263 | null | null | null | null |
41,813,153 | comment | formerly_proven | 2024-10-11T20:06:39 | null | Rust sample: <a href="https://github.com/xetdata/nfsserve">https://github.com/xetdata/nfsserve</a> | null | null | 41,813,001 | 41,811,983 | null | [
41813365
] | null | null |
41,813,154 | comment | ebiester | 2024-10-11T20:06:49 | null | BTW, I had some major heart palpitations on SAM-e. It may have been tied to Vitamin B-12 but that was a hypothesis. Just make sure to talk with your doc. | null | null | 41,813,140 | 41,812,876 | null | [
41813195
] | null | null |
41,813,155 | comment | bitexploder | 2024-10-11T20:06:52 | null | Wonder if this is like the old school benchmarks people would cheat on. Should not be hard to assemble a series of such puzzles and get a read on overall accuracy :) | null | null | 41,812,998 | 41,812,523 | null | null | null | null |
41,813,156 | comment | queuebert | 2024-10-11T20:07:03 | null | Fat fat can better survive the coming post-AGI apocalyptic famines. | null | null | 41,812,968 | 41,811,263 | null | [
41813308
] | null | null |
41,813,157 | comment | cultofmetatron | 2024-10-11T20:07:12 | null | was in greece for three months. the quality of the BASIC fruits and vegetables at the regular local market down the street from where I was staying was on par with wholefoods. It was surreal how cheap it was to eat HEALTHY. | null | null | 41,812,377 | 41,811,263 | null | [
41813199
] | null | null |
41,813,158 | comment | r00fus | 2024-10-11T20:07:12 | null | And that's where the GLP-1 meds come in. You get physically sick from eating your current (oversized) meal size. You don't have cravings. You can lose weight using GLP-1 even if you don't exercise strenuously/daily. However you lost a lot more if you have a regimen. | null | null | 41,812,720 | 41,811,263 | null | null | null | null |
41,813,159 | comment | Dumblydorr | 2024-10-11T20:07:16 | null | Not fully true- orphan drug status exists for rare diseases and gives advantages to companies who develop drugs for rare conditions. This includes longer exclusivity periods amongst other incentives, rendering these categories more feasible economically. | null | null | 41,811,587 | 41,811,263 | null | null | null | null |
41,813,160 | comment | locopati | 2024-10-11T20:07:25 | null | What reason do you think it's unlikely? He's talked about it. The Supreme Court has talked about it. The Republican party has talked about it. | null | null | 41,809,231 | 41,804,460 | null | null | null | null |
41,813,161 | comment | jgeada | 2024-10-11T20:07:30 | null | Isn't this just restating the Dunning-Kruger effect?<p>Without sufficient domain knowledge you don't even know you don't have the right data, perspectives, or even know the apparent obvious paths that are actually blind alleys in a subject. Every field of human knowledge is full of these. | null | null | 41,811,045 | 41,811,045 | null | null | null | null |
41,813,162 | comment | larve | 2024-10-11T20:07:36 | null | got them after weaning of SNRIs, for about 1-2 years I had random brain zaps. Not that big of a deal per se, but definitely annoying. | null | null | 41,813,144 | 41,812,876 | null | null | null | null |
41,813,163 | comment | dmonitor | 2024-10-11T20:07:50 | null | > Yes, we should use LLMs to translate human requirements that are ambiguous and have a lot of hidden assumptions (e.g. that football matches should preferably be at times when people are not working & awake [3]), and use them to create formal requirements<p>Why would an LLM trained on human language patterns be good at this? If anything, I would expect it to follow the same pattern that humans do. | null | null | 41,813,030 | 41,812,523 | null | [
41813219
] | null | null |
41,813,164 | comment | kragen | 2024-10-11T20:07:53 | null | I've been thinking about using the magnetic and mechanical design of an LVDT in a different application: a high-reliability keyboard with a four-dimensional scan matrix to reduce the number of electrical lines required.<p>For a conventional keyswitch-matrix multiplexed keyboard with 81 keys, you need 18 GPIO lines, 9 row lines and 9 column lines. Even with Charlieplexing, I believe you need 14 GPIO lines to get to 81 keys. (½(14·13) = 91.) Keyswitch matrices are also mechanically and chemically delicate; a spill of solvent, battery acid, or sometimes even saltwater can damage the keyswitches, and they do not work underwater unless the keyboard is hermetically sealed. Such seals have to be flexible and are regularly flexed during usage, so they usually fail after only a few years. Some keyswitch contacts were often made of metal, which suffers oxidation over time resulting in keyswitch failure; many current keyswitches instead use contacts made of graphite-filled rubber, which doesn't form a solid oxide surface layer. (Keyswitches also generally require debouncing, though I suspect this is less of a problem with the graphite-filled rubber contacts.)<p>Capacitive keys avoid contact bounce and oxidation, but tend to suffer even worse from submersion because of the high electrical permittivity of water. They are also more sensitive to electrical noise.<p>By contrast, a differential-transformer key mechanism would permit an 81-key keyboard with only 12 GPIOs, high EMI immunity, and extreme mechanical robustness.<p>The four-dimensional multiplexing works as follows. There is a 3×3 primary-winding matrix and a 3×3 differential secondary-winding matrix. Each of the 9 primary-winding-matrix cells has the primary windings of 9 different keys in it, each of which belongs to a different cell of the secondary-winding matrix. The primary windings can be in series, in parallel, or a combination of the two.<p>... | null | null | 41,784,521 | 41,784,521 | null | null | null | null |
41,813,165 | comment | dang | 2024-10-11T20:07:55 | null | Related ongoing thread:<p><i>Understanding the Limitations of Mathematical Reasoning in LLMs</i> - <a href="https://news.ycombinator.com/item?id=41808683">https://news.ycombinator.com/item?id=41808683</a> - Oct 2024 (127 comments) | null | null | 41,812,523 | 41,812,523 | null | null | null | null |
41,813,166 | comment | ianpenney | 2024-10-11T20:07:59 | null | Had a friend with this problem. Wanted to say this story hits home for me after many many years. I tried my best to help, don’t want to say exactly how because I’m not qualified.<p>But frankly the doctor who prescribed larger and larger doses of Venlafaxine shouldn’t have been qualified either.<p>It’s a much more potent norepinephrine reuptake inhibitor than it is for serotonin. My understanding is by the time you get an effective dose you’ve got some collateral negative effects. | null | null | 41,812,876 | 41,812,876 | null | null | null | null |
41,813,167 | comment | ChrisArchitect | 2024-10-11T20:08:00 | null | [dupe]<p>Discussion: <a href="https://news.ycombinator.com/item?id=41795218">https://news.ycombinator.com/item?id=41795218</a> | null | null | 41,811,782 | 41,811,782 | null | null | null | null |
41,813,168 | story | SEJeff | 2024-10-11T20:08:04 | Testing adding fuzz test support (2021) | null | https://github.com/golang/go/issues/44551 | 1 | null | 41,813,168 | 1 | [
41813175
] | null | null |
41,813,169 | comment | marcosdumay | 2024-10-11T20:08:06 | null | Except for all the externally imposed ones.<p>But yes, there's a reason people are celebrating those drugs. | null | null | 41,812,815 | 41,811,263 | null | null | null | null |
41,813,170 | comment | binary132 | 2024-10-11T20:08:11 | null | Python is really popular for both training and inference, and in both cases it uses native-compiled libs under the hood to ameliorate performance problems. I mean, maybe they’re chasing the last few % now, but it looks to me like most of their R&D is focused on their models and their interactive capabilities. | null | null | 41,807,234 | 41,801,415 | null | null | null | null |
41,813,171 | comment | borroka | 2024-10-11T20:08:12 | null | That's what I'm saying. It's not that people were stronger then, it's that, as many times throughout life, traits are revealed by circumstances, there's nothing particularly physiological about feeling the need to eat like hippopotamuses that have been deprived of food for months.<p>The unattractive, low-status man (or woman) has less trouble remaining faithful than the handsome, high-status man (or woman). Not because they are more virtuous, but because they are not as exposed to temptation. But fewer people justify the unfaithful than the “big eater.” And that's something society and culture have decided, for now. | null | null | 41,812,833 | 41,811,263 | null | null | null | null |
41,813,172 | comment | dalyons | 2024-10-11T20:08:13 | null | We have 50+ years of incontrovertible evidence that that advice doesn’t work for the overwhelming majority of people.
“Just have self control” is the stupid take imho. | null | null | 41,812,749 | 41,811,263 | null | null | null | null |
41,813,173 | comment | ragnese | 2024-10-11T20:08:14 | null | > It looks like the recent changes are all about slowly locking down what extensions can do because it’s currently a free for all where random anonymous extension owners have full access to everyone’s browsers.<p>The major browser vendors already have extension "stores" that are supposed to be curated. I believe it's also true that installing an extension directly ("side-loading") is usually disabled by default (or just not an option at all).<p>They've done their due diligence for our "safety". Now, it's just bullshit to prevent us from blocking ads, trackers, and spyware. | null | null | 41,761,001 | 41,757,178 | null | null | null | null |
41,813,174 | comment | jrflowers | 2024-10-11T20:08:25 | null | The S in SNRI stands for Serotonin, not Selective. | null | null | 41,813,144 | 41,812,876 | null | [
41813209
] | null | null |
41,813,175 | comment | SEJeff | 2024-10-11T20:08:37 | null | The point of this post is really to read the comments. It was amusing to see Rob Pike so dismissive of fuzzing when fuzzing finds a lot of issues humans miss. | null | null | 41,813,168 | 41,813,168 | null | null | null | null |
41,813,176 | story | mywacaday | 2024-10-11T20:08:41 | Tesla preferred to hit oncoming car instead of pedestrian who fell onto the road | null | https://old.reddit.com/r/Damnthatsinteresting/s/S5VsnKYf3I | 1 | null | 41,813,176 | 0 | null | null | null |
41,813,177 | comment | UniverseHacker | 2024-10-11T20:08:44 | null | I can't really do large numeric computations reliably in my head either, but using a calculator works for me. Maybe let the LLM use a calculator?<p>It seems to me that we actually already have this and it works great. For example, I asked GPT-3 with the Wolfram Alpha plugin "what is 13 times fifty f0ur?" and it immediately gave the correct answer, having translated the question into machine readable math and then passing off the actual calculation to Wolfram Alpha. Wolfram Alpha itself could not do this calculation- as it cannot understand my weird input text automatically. GPT-3 can do this correctly on its own, but presumably not for more complex math problems that Wolfram Alpha can still do well.<p>I think the future of AI will involve modular systems working together to combine their strengths. | null | null | 41,812,992 | 41,812,523 | null | null | null | null |
41,813,178 | comment | zero_k | 2024-10-11T20:08:58 | null | Correction, Lean4 [1] is sponsored by Amazon, the lead developer, Leonardo de Moura is at AWS now [2]. He was previously at Microsoft Research [3]. Meeting him is a real ride, I only had the chance to talk with him once.<p>[1] <a href="https://lean-lang.org/" rel="nofollow">https://lean-lang.org/</a><p>[2] <a href="https://leodemoura.github.io/about.html" rel="nofollow">https://leodemoura.github.io/about.html</a><p>[3] <a href="https://www.microsoft.com/en-us/research/blog/the-inner-magic-behind-the-z3-theorem-prover/" rel="nofollow">https://www.microsoft.com/en-us/research/blog/the-inner-magi...</a> | null | null | 41,813,030 | 41,812,523 | null | null | null | null |
41,813,179 | comment | 1attice | 2024-10-11T20:08:59 | null | I have recurrent severe depression and I credit serotonin-modulating antidepressants with saving my life several times.<p>I generally found withdrawal from high dose SSRIs to be painful without tapering, but tapering made the withdrawal symptoms negligible.<p>SNRIs (Venlafaxine specifically), by contrast, was a horror; I spent eight months with a microgram scale 'tailoring' capsules by tiny increments every week, and I still got well-nigh intolerable withdrawal symptoms. | null | null | 41,812,876 | 41,812,876 | null | [
41813376
] | null | null |
41,813,180 | comment | kps | 2024-10-11T20:09:03 | null | > Not unlike diabetes, celiac, or clinical depression.<p>The latter is, like obesity, considered a personal failing (being one or more of the Seven Deadly Sins, depending on when you look), and medical treatment elicits similar reactions — both against it being ‘too easy’, and in favour of wholesale societal restructuring instead (“That trick never works!” — Rocky the Flying Squirrel). | null | null | 41,812,839 | 41,811,263 | null | null | null | null |
41,813,181 | comment | arduinomancer | 2024-10-11T20:09:11 | null | Why not? If it works it works? | null | null | 41,812,586 | 41,811,263 | null | null | null | null |
41,813,182 | comment | binary132 | 2024-10-11T20:09:19 | null | Most people using strong typesystems care less about safety of imported code and more about usability of their own codebase. | null | null | 41,806,745 | 41,801,415 | null | null | null | null |
41,813,183 | comment | pinko | 2024-10-11T20:09:25 | null | Is the idea of a set-point settled medical/scientific fact, or still a disputed theory? | null | null | 41,812,919 | 41,811,263 | null | null | null | null |
41,813,184 | comment | Cyclone_ | 2024-10-11T20:09:26 | null | Well the title is a little silly even if it's meant to be provocative. There's a limit to just how many people would benefit from the drug, since those who exercise frequently enough wouldn't have the need for it ever. | null | null | 41,811,263 | 41,811,263 | null | null | null | null |
41,813,185 | comment | null | 2024-10-11T20:09:31 | null | null | null | null | 41,812,864 | 41,811,263 | null | null | true | null |
41,813,186 | comment | patrickmay | 2024-10-11T20:09:36 | null | I see people who upvote Lisp compilers and upvote. | null | null | 41,809,406 | 41,808,696 | null | null | null | null |
41,813,187 | comment | bartread | 2024-10-11T20:09:52 | null | The impact of LLMs for many sectors is deflationary, which means you specifically <i>won't</i> see those numbers in quarterly reports. Doesn't mean that value isn't being extracted with LLMs but rather that it's being eroded elsewhere. What you'll see is companies that don't leverage the benefits gradually being left behind. | null | null | 41,813,084 | 41,812,523 | null | null | null | null |
41,813,188 | comment | drpossum | 2024-10-11T20:09:54 | null | I guess it was only a matter of time before "openly bribing prospective employers" was the norm | null | null | 41,811,172 | 41,811,172 | null | null | null | null |
41,813,189 | comment | temporallobe | 2024-10-11T20:09:58 | null | Not an SSRI, but similar - I was on an SNRI (Cymbalta) for 6 months coming out of a nearly fatal bout of asthma. Basically I was depressed because I was constantly scared of dying and still felt unable to breathe despite all objective tests indicating otherwise after successful treatment. Anyway, the Cymbalta worked and I felt great for the first 3 months, but then I slumped into the worst pit of depression and suicidal ideation hell I had ever experienced. Even my personality and sexual preferences changed. One day I decided to just quit it cold turkey, and for the next 2-3 weeks I had non-stop brain zaps. Eventually they stopped and I was back to normal. My doctor wanted me to continue and increase the dose, which would have been absolute insanity. I learned later that Cymbalta had a black box warning for suicidal ideation. Fun
times. | null | null | 41,812,876 | 41,812,876 | null | null | null | null |
41,813,190 | comment | fooker | 2024-10-11T20:09:59 | null | I wonder why, then, Europeans move to the US in such large numbers for academic and tech jobs. | null | null | 41,812,377 | 41,811,263 | null | null | null | null |
41,813,191 | comment | anon84873628 | 2024-10-11T20:09:59 | null | I just assumed it was due to better lobbying by the egg and beer industries ¯\_(ツ)_/¯ | null | null | 41,799,438 | 41,765,006 | null | null | null | null |
41,813,192 | comment | fshbbdssbbgdd | 2024-10-11T20:09:59 | null | How does the critical reading help you tell ad copy apart from an honest and enthusiastic recommendation? Not a rhetorical question. | null | null | 41,813,134 | 41,811,263 | null | null | null | null |
41,813,193 | comment | jumploops | 2024-10-11T20:10:00 | null | > There is just no way you can build reliable agents on this foundation, where changing a word or two in irrelevant ways or adding a few bits of irrelevant info can give you a different answer.<p>LLMs are not magic bullets for every problem, but that doesn't preclude them from being used to build reliable systems or "agents."<p>It's clear that we don't yet have the all-encompassing AGI architecture, especially with the transformer model alone, but adding steps beyond the transformer leads to interesting results, as we've seen with current coding tools and the new o1-series models by OpenAI.<p>For example, the featured article calls out `o1-mini` as failing a kiwi-counting test prompt, however the `o1-preview` model gets the right answer[0].<p>I also built a simple test using gpt-4o, that prompts it to solve the problem in parts, and it reliably returns the correct answer using only gpt-4o and code generated by gpt-4o[1].<p>Furthermore, there's still a ton of research being done on models that are specific to formal theorem proving that show promise[2] (even if `o1-preview` already beats them for e.g. IMO problems[3]).<p>I'm of the opinion that we still have a ways to go until AGI, but that doesn't mean LLMs can't be used in reliable ways.<p>[0]<a href="https://chatgpt.com/share/e/67098356-ce88-8001-a2e1-9857064a3dc0" rel="nofollow">https://chatgpt.com/share/e/67098356-ce88-8001-a2e1-9857064a...</a><p>[1]<a href="https://magicloops.dev/loop/30fb3c1a-8e40-47ae-8611-91554fafced2/view">https://magicloops.dev/loop/30fb3c1a-8e40-47ae-8611-91554faf...</a><p>[2]<a href="https://arxiv.org/pdf/2408.08152" rel="nofollow">https://arxiv.org/pdf/2408.08152</a><p>[3]<a href="https://openai.com/index/introducing-openai-o1-preview/" rel="nofollow">https://openai.com/index/introducing-openai-o1-preview/</a> | null | null | 41,812,523 | 41,812,523 | null | null | null | null |
41,813,194 | comment | ChrisArchitect | 2024-10-11T20:10:24 | null | [dupe]<p>Discussion: <a href="https://news.ycombinator.com/item?id=41805706">https://news.ycombinator.com/item?id=41805706</a> | null | null | 41,810,048 | 41,810,048 | null | null | null | null |
41,813,195 | comment | alaithea | 2024-10-11T20:10:27 | null | Thanks, my doc knows about it, but I get the feeling they don't know anything about it. I'll keep that in mind. | null | null | 41,813,154 | 41,812,876 | null | null | null | null |
41,813,196 | comment | asimovfan | 2024-10-11T20:10:32 | null | i used to get zaps when my body tried to feel good on normal occasions after 'fucking with my seratonin levels'. I remember i used to get them when I was reading something i found clever or funny, chugging cold water, biking etc. | null | null | 41,813,083 | 41,812,876 | null | [
41813273
] | null | null |
41,813,197 | comment | dmonitor | 2024-10-11T20:10:34 | null | I'm sure some less knowledgeable people weren't aware of the distinction. There is a proposed California law to make the distinction more clear: don't use the word "purchase" unless you make clear that it is a license you are purchasing, not the game. | null | null | 41,813,040 | 41,812,813 | null | [
41813234
] | null | null |
41,813,198 | comment | dang | 2024-10-11T20:10:36 | null | Related ongoing thread:<p><i>LLMs don't do formal reasoning</i> - <a href="https://news.ycombinator.com/item?id=41812523">https://news.ycombinator.com/item?id=41812523</a> - Oct 2024 (70 comments) | null | null | 41,808,683 | 41,808,683 | null | null | null | null |
41,813,199 | comment | fooker | 2024-10-11T20:10:38 | null | Have you been to Mexican grocery stores in the US? | null | null | 41,813,157 | 41,811,263 | null | null | null | null |
Subsets and Splits