Putting aside the specifics for a second, I'm sorry to hear about your injury and glad you've found workarounds. I also think high-quality voice transcription might end up being a big thing for my health (there's no way typing as much as I do, in the positions I do, is good).
Much appreciated. I find is that referencing code in conversation is hard -- e.g. "underscore foo bar" vs `_fooBar`, "this dot Ls" vs `this.els`, etc happens often. Lower-powered models especially struggle with this, and make some frustrating assumptions. Premium models do way better, and at times are shockingly good. They just aren't remotely economically viable for me.
My solution so far is to use my instructions to call out the fact that my comments are transcribed and full of errors. I also focus more on "plan + apply" flows that guide agents to search out and identify code changes before anything is edited to ensure the relevant context (and any tricky references) are clearly established in the chat context.
It's kinda like learning vim (or emacs, if you prefer). First it was all about learning shortcuts and best practices to make efficient use of the tool. Then it was about creating a good .vimrc file to further reduce the overhead of coding sessions. Then it was about distributing that .vimrc across machines (and I did a LOT of ssh-based work) for consistency. Once that was done, it became unimaginable to code any other way.
It has been even more true here: agent-based workflows are useless without significant investment in creating and maintaining good project documentation, agent instructions, and finding ways to replicate that across repos (more microservice hell! :D) for consistency. There is also some conflict, especially in corporate environments, with where this information needs to live to be properly maintained.
maybe you've done this already, but my first thought would be to make a preparser script that would take your likely voice inputs like "underscore foo bar" and translate to "_fooBar" which you would then pass on as input. i do something similar for a local TTS generator which often stumbles on certain words or weird characters
> Something that surprised us early on: only a tiny fraction of farmed fish species have been through genetic improvement programs. Chickens grow 4x faster than they did in 1950 because of decades of selective breeding.
I agree that there is an opportunity here for getting more calories per fish (and especially per input of feed, which is really what decades of chicken optimization are about). But the consequences of these changes for chicken welfare have been disastrous [0] and we're seeing a concerted effort to move to higher-welfare breeds (though still more efficient than ancestral breeds). Likewise, intensive salmon farming has led to widespread '“environmental dewilding,” or the process of modifying natural water bodies with artificial infrastructure — in this case, fish farm pens and cages — and polluting them' [1]. It sounds like there are lots of ways in which using more robots can make monitoring less-invasive, and therefore less stressful for fish. I certainly hope to see those attributes, rather than the potentially disastrous ones, emphasized as you move forward.
This is a really important point and something we think about a lot. You are absolutely right that chicken optimization has come with serious welfare tradeoffs. Breeding purely for growth rate without considering the animal's ability to actually live comfortably is how you end up with birds that cannot walk properly. We do not want to enable that trajectory for fish.
The good news is that the data we are collecting can be used to select for more than just growth. Body shape, fin integrity, spinal curvature, and other morphological traits are all signals of fish health and welfare. Farms that care about sustainability can use this data to breed fish that are robust and healthy, not just fast growing. The tool is neutral but the selection criteria are up to the breeder.
On the environmental side, our focus right now is on land based hatcheries and recirculating aquaculture systems rather than open net pens in the ocean. These closed systems avoid a lot of the dewilding and pollution concerns you mentioned. They are more expensive to operate but they keep farmed fish separate from wild populations and give you much more control over waste and water quality.
And yes, reducing handling stress is a big part of what we are building. The manual process today involves netting, anesthesia, and physical manipulation. Our system can measure fish without any of that. Less stress on the animal and better data for the farmer.
> Let’s say every company gets about three innovation tokens. You can spend these however you want, but the supply is fixed for a long while... If you choose to write your website in NodeJS, you just spent one of your innovation tokens. If you choose to use MongoDB, you just spent one of your innovation tokens. If you choose to use service discovery tech that’s existed for a year or less, you just spent one of your innovation tokens. If you choose to write your own database, oh god, you’re in trouble.
From my POV, the author spent their innovation tokens on a political commitment. I would not recommend this path to someone starting a company. It's hard enough already.
Also, many American companies that might have been useful to the author were founded by Europeans, e.g. GitLab. There's plenty of European talent for making widely adopted infrastructure. If those companies aren't in Europe, it's worth asking why [1].
But the argument is reversed! The more boring your tech stack, the _easier_ it is to host it anywhere (including Europe). So choosing boring tech is actually an enabler of this (and other) choices down the line.
It's only "a political commitment" as long as it doesn't affect you yet; and from the European perspective I'd say "the affecting has begun".
I'd say from this author's POV, his commitments cost him in terms of headaches, costs, and time not spent optimizing for meeting customers' needs:
> The parts that were extra hard
> Transactional email with competitive pricing. This one surprised me. Sendgrid, Postmark, Mailgun, they all make it trivially easy and reasonably cheap.
The EU options exist, but finding one that matches on deliverability, pricing, and developer experience took real effort. Scaleway's TEM works, but the ecosystem is thinner. Fewer templates, fewer integrations, less community knowledge to lean on when something goes wrong.
The choose boring technology essay notes that as you get further along you might get more innovation tokens to spend. but when you're starting out, "not choosing sendgrid because they're American" is a token gone when they're most scarce.
Fair enough... though if I were to push my point: one could also say that dumbing down your mechanisms of email sending (i.e. ditching templates, or pulling the templates to your own codebase) would give the same advantage I talked about earlier of vendor-independance
> Can you find an academic article that _looks_ legitimate -- looks like a real journal, by researchers with what look like real academic affiliations, has been cited hundreds or thousands of times -- but is obviously nonsense, e.g. has glaring typos in the abstract, is clearly garbled or nonsensical?
It pointed me to a bunch of hoaxes. I clarified:
> no, I'm not looking for a hoax, or a deliberate comment on the situation. I'm looking for something that drives home the point that a lot of academic papers that look legit are actually meaningless but, as far as we can tell, are sincere
Close, but that's been retracted. So I asked for "something that looks like it's been translated from another language to english very badly and has no actual content? And don't forget the cited many times criteria. " And finally it told me that the thing I'm looking for probably doesn't exist.
For my tastes telling me "no" instead of hallucinating an answer is a real breakthrough.
> For my tastes telling me "no" instead of hallucinating an answer is a real breakthrough.
It's all anecdata--I'm convinced anecdata is the least bad way to evaluate these models, benchmarks don't work--but this is the behavior I've come to expect from earlier Claude models as well, especially after several back and forth passes where you rejected the initial answers. I don't think it's new.
I can concur that previous models would say "No, that isn't possible" or "No, that doesn't exist". There was one time where I asked it to update a Go module from version X.XX to version X.YY and it would refuse to do so because version X.YY "didn't exist". This back with 3.7 if I recall, and to be clear, that version was released before its knowledge cut off.
I wish I remembered the exact versions involved. I mostly just recall how pissed I was that it was fighting me on changing a single line in my go.mod.
alas, 4.5 often hallucinates academic papers or creates false quotes. I think it's better at knowing that coding answers have deterministic output and being firm there.
It would have to have been trained on the papers without being aware of retractions for that test to work. Otherwise it will be limited to whatever papers it gets from a search engine query, which likely won't contain any un-retracted illegitimate papers.
I once worked at a crypto company that outsourced some of its marketing work to a content marketing firm. A piece that firm submitted to us contained a link to an "academic" article about global poverty with a totally garbled abstract and absolutely no content whatsoever. I don't know how they found it, because when I search google scholar for a subject, usually the things that come back aren't so blatantly FUBAR. I was hoping Claude could help me find something like that for a point I was making in a blogpost about BS in scientific literature (https://regressiontothemeat.substack.com/p/how-i-read-studie...).
The articles it provided where the AI prompts were left in the text were definitely in the right ballpark, although I do wonder if chatbots mean, going forward, we'll see fewer errors in the "WTF are you even talking about" category which, I must say, were typically funnier and more interesting than just the generic blather of "what a great point. It's not X -- it's Y."
When Claude does WebSearch it can delegate it to a sub agent which of it ran in the background will write the entire prompt on a local file and the results. If that happened, I would like to know what it gave you for that. It is always very interesting to know the underlying "recall" of such things. Because often it's garbage in garbage out.
The location might still be on your disk if you can pull up the original Claude JSOn and put it through some `jq` and see what pages it went through to give you and what it did.
> The user wants a real, sincere academic paper that looks legitimate — proper journal, real authors with real affiliations, many citations — but is obviously nonsense or garbled. Not a hoax or deliberate prank, but something that was genuinely submitted and published, yet is clearly meaningless if you actually read it.
Let me think about what I know and search for examples. There are some famous cases:
> Papers generated by AI/GPT that slipped through - there have been reports of papers with phrases like "Certainly, here is a possible introduction for your topic" or "As an AI language model..."
> There have been papers with obvious tortured phrases - "sham" papers that use synonym substitution to avoid plagiarism detection, producing phrases like "sheep pursuing" instead of "shepherd" or "big data" becoming "large information"
I'm also thinking of papers that made it through peer review despite being complete gibberish, like the famous SCIgen computer-generated papers that fooled conferences, or papers where the abstract and content don't match at all.
> Now I'm trying to recall specific real examples that fit this criteria — published papers that are genuinely incoherent or nonsensical but appeared in legitimate venues.
This is maybe not the _exhaustive_ search strategy I would have liked, but it's probably something you would do if you were limited to just a certain amount of effort on a task.
It might be wrong but that’s not really a hallucination.
Edit: to give you the benefit of doubt, it probably depends on whether the answer was a definitive “this does not exist” or “I couldn’t find it and it may not exist”
claude said "I want to be straight with you: after extensive searching, I don't think the exact thing you're describing — a single paper that is obviously garbled/badly translated nonsense with no actual content, yet has accumulated hundreds or thousands of citations — exists as a famous, easily linkable example."
That's still less leaned toward blatant lies like "yes, here is a list" and a doomacroll size of garbage litany.
Actually "no, this is not something within the known corpus of this LLM, or the policy of its owners prevent to disclose it" would be one of the most acceptable answer that could be delivered, which should cover most cases in honest reply.
> no, I'm not looking for a hoax, or a deliberate comment on the situation. I'm looking for something that drives home the point that a lot of academic papers that look legit are actually meaningless but, as far as we can tell, are sincere
The Sokal paper was a hoax so it doesn’t meet the criteria.
Results from a one-shot approach quickly converge on the default “none found” outcome when reasoning isn’t grounded in a paper corpus via proper RAG tooling.
Can you provide more context to your statement? Are you talking about models in general? Or specific recent models? I'm assuming "one-shot approach" is how you classify the parent comment's question (and subsequent refined versions of it).
I’m glad to hear that. Another frame is that your depression turned out to be “math hard” rather than bodybuilding hard [0]. Your disciplined, methodical approaches were steady applications of effort, whereas what you actually needed was easy to implement but hard to envision.
We are indeed living in more comfortable homes. Americans are migrating to the sunbelt because of ample AC in the summer and the winters are pleasant. that’s a big part of why we have many fewer heat deaths per capita than Europe: https://www.thetimes-tribune.com/2025/08/02/opinion-us-heat-...
You don’t realize how nice it is to live and work in air conditioned spaces until visiting a part of Europe where AC is viewed with disdain for reasons I still don’t understand.
Also the move to electric heat pumps is increasing electricity rates but reducing natural gas usage and improving overall efficient.
The GP comment was trying to do snarky doomerism but accidentally hit upon a lot of truths. It’s amazing how many things are getting better but some people are hell bent on being cynical about it anyway.
> You don’t realize how nice it is to live and work in air conditioned spaces until visiting a part of Europe where AC is viewed with disdain for reasons I still don’t understand.
Most of Europe is poor. AC is expensive. It's actually that simple.
I have lived and worked in Switzerland. My office (shared with 2 other people) was the only space in the entire floor with AC due to some obscure archaic reason.
That air conditioning worked great for years, but a few months before I left that position, the facilities management people suddenly came in and ripped it out. No justification given.
Thank God TPTB didn't notice I had AC for all those years; it really would have been miserable without it. But despite the misery I noted all around me, there was an extremely strong disdain for air conditioning that permeated the culture. When I talked to friends and colleagues about the AC situation I was regularly ribbed for being a gluttonous American wasting electricity on such a triviality. They were legitimately proud to suffer. Baffling.
I've come to the conclusion that most Western and Central Europeans--yes, including Swiss--have a masochistic superiority complex around AC. They see suffering without AC as core to the European identity and sweating it out in unproductive misery (or taking a whole month off of work) as virtuous. They willingly kill thousands of people and leave hundreds of millions more in misery every year simply to feel superior and European.
Not at all, it has one of the lowest rate in Europe along with the UK. It's very hard to get the building permit required to install one.
Portable AC has had a boom those past few years though (because it doesn't require a permit).
Europe is so backwards when it comes to annual heat deaths that they manage to have more heat deaths per year than the US has gun deaths + heat deaths combined. You won't hear about that from Europeans though, it'd make them seem barbaric. 175,000 heat deaths per year in Europe according to the WHO. It's a staggering genocide of technological primitiveness. Imagine having millions of people die because you can't be bothered to adopt 1950s technology (and of course I'm aware of the things the US is backwards on).
I think it is simply because in most of Europe air conditioning is unnecessary for comfort 95% of the year. Here in San Francisco most homes don't have air conditioning either, but there might be a week or two where it gets very hot and we just put up with the barbaric technological primitiveness.
Much of the US is extremely unpleasant without air-conditioning for a substantial portion of the year so of course everyone living in those parts installs it.
Yes in many parts of the US it's not just the heat but the humidity. One summer I tried going without AC as much as possible to see how much it would change my electric bill. I could handle the temperature most of the time but the humidity especially at night started giving me mold problems in the house. Cleaned that up and went back to using the AC and no more mold. Not sure how people controlled this back in the pre-AC days, maybe just a lot more cleaning.
My parents still don’t use AC. The windows stay open all summer unless there is a rainstorm. Whole house fan is turned on at night to draw in cooler air. Much time spent in the cooler basement if you are going to be hanging at home. At night you are basically sleeping naked on top of your fitted sheet with one or two window unit fans circulating air. Maybe another fan pointed directly at you. Basement had some dehumidifiers and afaik that was the only problem moisture area.
You sent me to the books because this is such a fascinating stat. It's true! Heat deaths in the US: 5 per million people. Italy: 500+ per million people. I had no idea.
Figures based on coroners reports are somewhat suspect.
> In September 2022, a vicious heat wave enveloped much of the western U.S., placing tens of millions of people under heat advisories. Temperatures across California soared into the triple digits. Sacramento broke its heat record by more than 6 degrees Fahrenheit when the temperature hit 116 degrees.
> California death certificates showed that 20 people died as a result of heat-related illness from Aug. 31, 2022 to Sept. 9, 2022.
> But a study last year by California’s Department of Public Health found that death rates increased by about 5 percent statewide during the heat wave, causing 395 additional deaths.
Thanks for the paper link, very different figures from the random USA newspaper article :)
I'd love to see an age adjusted figure as well as it's likely Europe has likely more very old people and my guess is that heat/cold mortality is concentrated in the very old people.
I suspect age distributions are part of the story. Also the Eastern US (where most of the population lives) experiences much larger swings in temperature between winter and summer so maybe people are just more prepared for it.
It's not the most convenient format because of their idea of what constitutes a region, but yeah, the US has a pyramid shaped population pyramid, while European regions have a big bulge of old people: https://www.cia.gov/the-world-factbook/references/population...
I do think we'll need to change our view on airconditioning, every home should have airconditioning just like it has heating.
But I'm very sceptical of those numbers. They are apparently even worse for cold, and you can't attribute that to lack of airconditioning. I still think the huge difference can only be attributed to a difference in reporting.
I think it’s common knowledge China never reports numbers that cast them in an unfavorable light, to the point where analysts generally disregard them.
It's not as simple as it might seem at first glance. People often go into their basement and think "wow, it's cool down here. If only I could make my house this cool." But, as soon as you moved the air from your basement to your house, the air in your basement would be replaced by ambient air and would take time to be cooled by the Earth. And so you quickly realize you need a lot of thermal mass and an efficient way to move heat in order to keep up with removing the heat from your house.
Yes, but it needs to work both ways. Heat needs to be extracted during the winter. Otherwise the ground would just be heated up to much. That is what a ground source heat pump does.
I think there wasn't a culture of buying ACs, because in most of Europe the climate was much more moderate. The summers are much hotter now than when I was a kid and heat waves are more regular. Many more people are buying air conditioning now.
Much of the US already had warmer summers than Europe when the impact of climate change was smaller, so AC is far more common.
I'm as ready to hate on Meta as anyone but this article is a bit of a nothingburger.
So there are disagreements about resource allocation among staff. That's normal and healthy. The CEO's job is to resolve those disagreements and it sounds like Zuck is doing it. The suggestion to train Meta's products on Instagram and Facebook data was perfectly reasonable from the POV of the needs of Cox's teams. You'd want your skip-level to advocate for you the same way. It was also fine for AW to push back.
>. On Thursday, Mr. Wang plans to host his annual A.I. holiday party in San Francisco with Elad Gil, a start-up investor...It’s unclear if any top Meta executives were invited.
Egads, they _might_ not get invited to a 28-year-old's holiday party? However will they recover??
Can somebody explain to me how giving a 28 year old kid 250 million (or was it 1 billion) to run your AI lab is a good idea? Or is it actually a dumb idea? I think it is a dumb idea, but maybe somebody can make it make sense.
well if the expected value of developing AGI is 100 quadrillion dollars -- 1000X bigger than the entire global economy -- and you think this person has a .01% chance of getting there in any given year, you should pay him 10 trillion dollars a year :)
I think a surprisingly good AI will come into existence and one of the lessons will be that we greatly overvalue intelligence over basic distribution. Giving a single kid millions and billions is symbolic of the actual problem (the distribution one).
Also, there's basically 0% chance this kid is one of the top 1000 most knowledgeable people in the world on this technology.
The other possible future is you rent the car for exactly when you need it and don’t pay a monthly bill— or your monthly bill pays for a certain number of rides/minutes/miles per month. In which case the subscription costs are managed by the provider, who might be the manufacturer and might not.
At least in cities, a fully-functioning, on-demand autonomous fleet would probably be superior to car ownership in just about every way except as a status symbol.
The monopolist providing this service would be de-incentivized from ever equipping for all the demand, and the last 10% of capacity being bid on by the last 20% of demand would make this a constant stress and struggle.
Meanwhile it's an excuse for another century of more car lanes and less mass transit infrastructure.
There used to be a service like this, called Car2Go. Not autonomous, but more like how scooter/bike rentals work. It was fantastic, and in no way profitable.
Deep Sea one is scary for some reason. It just gives me shivers to think about how deep the sea is, and what horrors lurk down there. I know that I'll never encounter such a being, but still kinda creepy.
I love Neal's work so much. He's constantly making some of the coolest stuff on the web. I'm utterly delighted every time I see his domain on the front page of HN.
I hope he never stops making these art pieces - everything he creates brings joy, regardless of whether it's educational or funny or whimsical. I'm in awe of his creative output, his manner of communication, and his ability to steal hours of our time playing ridiculous little games that make us question the fundamentals of life and society.
He's right up there with XKCD in my mind.
--
This is probably the only time I'll use my super pedantic mode on Neal's work, and it's only because I love biology -
> DNA
> The genetic instructions for life
> 3.5 nanometers tall
DNA has a lot of dimensional metrics. It gets complicated. The people that study this stuff really care because it's essential for how our enzymes work, and small differences in spacing tolerances would totally break all of the machinery.
This "3.5 nm" figure is roughly the height of one turn of the helix for one form of DNA (B-DNA). The figure is showing multiple turns in the cartoon illustration.
In theory, you could create a polymer of infinite length (or height).
B-DNA is 34 Å per turn, with 10.5 bp per turn (table 1) :
> King of the animal kingdom, it is the largest animal to have ever lived. It can eat up to 40 million krill per day during peak feeding season.
Please fix this one, Neal! We don't know that the blue whale is the largest animal to have ever lived (even assuming we know we're just talking about earth).
Blue whales are perhaps the largest animal to have ever lived on earth. But we simply do not know. The fossil record is woefully incomplete.
We even have new papers coming up all the time that challenge this:
This is undoubtedly the last time the claim to largest will ever be challenged. Even if we dug up no new fossils, the estimations of previous finds change all the time as we learn more.
Also - what does "largest" mean? Mass? Length? Surface area?
It's okay to say that they're the largest (by some metric) that we know of. But it is not correct to say that they're the largest to have ever lived - at least as far as we know or could ever know. And by setting an absolute, inquiring minds memorize the point and stop wondering.
It's very probable that we'll never know the definitive answer to this.
> (even assuming we know we're just talking about earth)
This is a nitpick, but life on other planets wouldn't be called “animals”. Animal is a clade defined by common ancestry. The only way you could have an extraterrestrial animal is for it to have evolved on Earth and then migrated somehow, and I think we can fairly confidently rule that out.
The dimensioning of DNA was an immediate turn-off for me. A common biochemistry demo is to show how long and macroscopically visible a chromosome can be. Saying DNA is 3.5 nm tall (long?) flies in the face of what is a pretty interesting and notable experience for a lot of people.
It essentially starts the whole project with a weird take on "How long is a piece of string?"
> In theory, you could create a polymer of infinite length (or height).
