> Over the years, Pitelka has rescued some of these rocks and deposited them at his mom’s five-acre property, including a large hunk of limestone that, when he dumped it out of his truck, split in half to reveal a perfect trilobite fossil, convex on one side and a matching imprint on the other.

> “I was so excited to hear that the right people got involved,” he said. “I’m absolutely ecstatic it’s going to be on display. For me, as an (excavator) who loves boulders, it’s the coolest thing that people understand its historical significance and appreciate it.”

[0]: https://www.bbc.co.uk/news/uk-36225652

  • Sweet Stone Chicago
  • No ORDinary Rock
  • Chicago Rocks
  • O'Rock
  • Orchard Rock
  • Da Rock

So deciding between the two, Orchard Rock is my pick! It's subtle, but the plaque can explain it.

> I doubt much will be merged upstream. Many updates we have can technically run on other OSes (%gs stubs won't be built on Linux because %gs is correctly set, SIGUSR1 will work but less optimally, etc). However I picked an approach where everything that is done for OpenBSD is just marked as OpenBSD only. I did that purely to simplify my life while I was working on it and clearly mark the changes.

> When I was talking to Wine maintainers a month ago, they were hesitant to accept even patching Wine so it compiles for OpenBSD, mostly because it's another OS they will have to support and keep operational, while the %gs stub fix is something they definitely cannot take upstream. They are very hesitant to take anything that is changing the "emulated" application code.

1) Where possible, don't test in animals.

2) When testing in animals, first ensure that there is reasonable confidence the risk and severity of harm from the testing will be minimal. Only begin initial in vivo testing after this has first been achieved.

3) For test animals that are or aren't harmed, allow them to live happy, (relatively) normal, natural lives. Don't end their life early unnecessarily, with exceptions for humane euthanasia if the testing has unexpectedly caused significant, persistent suffering. Treat them as a pet owner would treat their beloved pet rat.

Some may consider it extreme, but, in my view, in an ideal world, people would treat testing in animals almost like testing in humans. The researcher should first ask "would I feel comfortable testing this in a voluntary human test subject?" Non-human animals would still be the first line of testing, since harm/death in a rat/mouse/chimpanzee is less morally bad than in a human, but I think this should be the mindset.

> In her appeal, Harpe said her husband was a certified federal law enforcement officer and that every search was conducted for law enforcement purposes.

> “The violation occurred because I gave my husband, who is a certified law enforcement officer with the Department of Homeland Security, access to my Flock account,” Harpe wrote. “Every search he conducted was for law enforcement use only and not for personal gain.”

There seems to be a subset of law enforcement officers who fundamentally don't understand privacy, checks and balances, and responsibility. They believe they should have unrestricted access to everyone else and that they are above us all.

Firing her was not excessive, it is the bare minimum. Officers should be held to a higher standard, not lower. She and her husband should be in a jail cell.

This is a great point, and one that really isn't played up because it doesn't fit the whole profit-from-attention narrative that literally every media surface seems to be following.

It almost seems as if law enforcement 'need' these additional tools to keep them busy; find new, more trivial crime that's been hiding behind no one reporting it or being hurt by it.

Maybe if law enforcement resources could be applied to more corporate crime, that'd keep em busy till the cows come home.

Many people pursue careers based on the intersection between their interests and what society considers valuable. Being upset when what they chose to spend their lives doing either loses value or changes so much it becomes a wholey different job is understandable, and calling it "rent seeking" is very unfair, in my view.

But why can't a hyper-intelegent AI also do this biology-esque math work of explaining its work to feeble humans?

It is easy for a genius to have theory of mind over an idiot but not vice-versa. So it should be easier for a genius AI to explain its work to an idiot human than for an idiot human to dissect the work of an genius AI.

Also, we're not talking about solutions that are arrived though an evolutionary/statistical process like explaining the design of an organism or the weights of an LLM. We are talking about solutions that are arrived through lingual/textual/rational processes like mathematics. So it's fundamentally a different mode of thinking.

It's not that we have "a brain to study", it's like if we had an "explaination of how a brain works" to study.

The task of breaking down these AI proofs is not translating and analyzing a fuzzy real-world system into a hard logical system (like in biology), but in translating a hard logical system into a simpler hard-logical system we can understand.

No biologist has ever complained, "Gee whiz, it's so unfortunate that we already have a brain to study. It's all so meaningless! It would be so much better if we had to construct the first-ever brain from scratch."

> For 33 hours in late August, a rival network hijacked a block of Hetzner-hosted infrastructure serving Softaculous and Virtualizor, using it to issue a valid TLS certificate and push a malicious software update. The stolen route was RPKI-invalid the entire time, on file, checkable by anyone. It still won on most of the internet.

> At 20:57 UTC on 28 August, a network nobody at Hetzner or Softaculous had dealt with before started announcing 162.55.80.0/24 to the internet. It was a small, unremarkable-looking slice of Hetzner’s own address space — the block hosting Softaculous’s client area, its billing pages, and the update endpoint that every Virtualizor-licensed control panel on the internet periodically calls home to. For the next 33 hours, with an 11-hour lull in the middle, a meaningful share of global traffic bound for that block went to the hijacker instead. Some of it was Let’s Encrypt’s own domain-control validation, which is how the attacker walked away with a certificate for virtualizor.com that no browser would ever flag.

* https://cloudpunk.blog/post/bgp-hijack-rpki-gap/

So it's all very well to have multiple network locations to prevent hijacking on the path (i.e., you have to attack multiple paths), but if the network as a whole thinks the destination is someplace else then all those paths will lead to the same (false) location.

Perhaps the CAB should consider RPKI validation as a "MUST" rather than piling on multiple sources of validation who still have the same single point of failure (the spoofing of the destination).

https://w2.eff.org/Misc/Publications/John_Perry_Barlow/HTML/...

My father is mostly on Apple and tries to use his iPad as a PC, ie with files and "the filesystem", he tries to get out of using extra iCloud for pictures (using Proton, or he asks me about Immich, or he wants to use an external HDD, or use Synology Photos and backup). He's mostly on the iPad + keyboard because it's a nicely managed device update-wise and "what I can install"-wise, virus-wise.

But I recent told him: Just give up. Buy (rent!) the storage, you are too depended on me, there is always something with the non-Apple solutions. Yes, your iTunes library is a mess on the filesystem, just don't look at it through anything else than "Music". Yes, Proton Drive eats 400 GB of your iPad storage now, I don't know how to trim that regularly. I'm sorry, it's either Linux and I can help set some things up, or full Apple. Choose. His current mixed, "trying to get best-of-both-worlds", approach is just pure unexpected pain, every time again.

Edit: I also have an iPad, it does work for me, but where my father is trying to use his iPad as a central, primary device (the hub), for me it is a client (spoke) of my own infra. Apps get access to information and I never use Apple services for anything. So indeed, Immich, Proton, Home Assistant, government and banking apps, etc. My iPad is throw-away: If it dies, it dies... I buy a new one, log into all (self-hosed) services and I'm 100% back up and running. Doing the same for my father is also pain, because his Home Assistant will have issues, or his Immich instance, and it's real work to fix, and he won't tell me for months if it does not work and work piles up. So, I tell him to surrender to Apple... It die a little inside.

I have found most of the other skills to emulate an 'enterprisy' way of developing, having everything checked multiple time by 'councils' and whatnot, making things slow, vacuum tokens like mad and the whole process reeks of cargo culting.

He probably pulls 20-100 [0] people over every day. All of those checks are unconstitutional. Once every 1-100 days one of them has something worth going to court over. Most of those will be plea-bargained and never reach a judge. If it's sufficiently serious, though, it's worth the fight in court. So the judge will only see the really serious cases.

And we're only seeing the reporting because the judge did something unusual and ruled the entire operation unconstitutional.

So it's not "wow! those cops really know their business; they pulled over the one person carrying meth!". It's "wow! out of thousands of illegal, unconsitutional, vehicle stops, one person got taken to court and put in front of a judge, and that case was thrown out"

[0] I have no idea on actual numbers. The point still holds.

In this case, the Flock data then showed a bunch of recent interstate travel, which cops consider to be probable cause. They know it’s not legally probable cause though, so they pointed to her “nervousness” (aka not consenting to search).

The real question is how many times the same cop pulled over people who he searched and found nothing. Certainly nonzero.

There is a common understanding with respect transformers the the output text is the only channel between one invocation and the next. ... so if it wanted to be thinking some 'hidden thoughts' it would have to be modulating its word-choice.

I think this is an artifact of tutorials on attention showing the system proceeding and looking back at words. But the information in the KV cache is not words-- it's entirely latent data, and in the interior layers they KV caches don't even tend to have that much to do with words. Their only guaranteed relationship with the input words is that they're a deterministic function of all the input tokens up to that point.

This means that a autogressive transformer could be having a whole complicated invisible discussion going on in the KV entries which was arbitrarily unrelated to what it was saying. And if it wasn't so it would be much more difficult do stuff like exactly copy a list but then truncate at the Nth entry as the simplest way to do that is to count down internally.

This is not, in my view, a concern or issue-- the model saying stuff that is inconsistent with it's "thinking" is a very high entropy configuration that would not be selected for in training unless you were more or less trying to do that. So not even 'could be' --but always is as an integral part of working.

And the useful safety relevant advice is don't undertake training that trains the model to be two-faced and it simply won't be... because being two-faced is a much more complicated state. You can even see it in humans w/ brainwashing: force someone to say something enough and they'll usually believe it, not just say it even though you were only supervising what they said.

Ironically and unfortunately the only credible kind of training I can think of that might produce duplicitous behavior is RL with "safety filters"--- if you have some dumb nanny process watching what the model says but then reward it for achieving a task, you're doing exactly the kind of training that can produce that mismatch. "To solve this task I need to hack this process but if I do so obviously I'll get punished..."

The greater concern with latent reasoning would be that potentially you'd be operating in a regime where the total computational depth is many times the model*kvcache size hold, potentially in a regime that was never trained at all. It could be arbitrarily far out of distribution. It might even say a naughty word or tell a racist joke!

If OpenAI had never made a push to get good at maths, we might still have a sizeable group of peopne stuck at that first line. I say that because I still hear in discussions the naive rebuttal "But we'll still need _someone_ to {one-step-up-in-hierarchy}" . Although you have to be curious now how long that sentiment can last.

The good news for those doing abstract maths might be that, now that OpenAI have proven that their AI can do the job in principle, there's probably not much left in it for them to keep going. Mathematicians should know better than anyone the lack of a direct profit motive in doing the work that they do.

(Edited for clarity)

He described it at 'strip mining', 'rude', and said that many grad students, who are given problems like these as a part of their training, had the rug pulled out from under them (because its customary to let people know you are working on something to prevent this kind of thing).

And your whole analysis is that Tao is worried about his 'identity', status, and 'being a differentiator'.

This is projection on your part. People have values that bear no resemblance to yours and it would behoove you to learn to recognize that.

Consider any man at his job, fighting the good fight. Your boss says "I've hired this man to assist you" and you are happy to finally have some help with your difficult task. Then one morning you come into work and see that the coming day’s work was done overnight by this other fellow. And the realization sinks in, he isn't there to help you, he is there to replace you.

In this view his reaction seems totally expected. I expect many professionals, e.g. doctors, lawyers, accountants, etc. will make this realization sooner or later. The companies are saying these tools will assist professionals, but in reality they are aiming to replace them.

  Temba, his arms wide stdio
  
  Darmok on the ocean
      Kira at Bashi "Hello, World!\n"

      Darmok and Jalad at Tanagra 0
  Darmok and Jalad on the ocean

This, unfortunately, takes time, as one has to restart many times because of ideas that turned out not to be as great as one had initially thought. And it results in software that is called "opinionated", and often does not follow what the "industry" does.

Evidence of the bouba-kiki effect in naïve baby chicks - https://news.ycombinator.com/item?id=47105198 - Feb 2026 (107 comments)

Which is "Bouba", and which is "Kiki"? [video] - https://news.ycombinator.com/item?id=46572742 - Jan 2026 (23 comments)

Bouba / Kiki Effect - https://news.ycombinator.com/item?id=45320327 - Sept 2025 (2 comments)

Bouba/kiki effect - https://news.ycombinator.com/item?id=40699583 - June 2024 (170 comments)

Bouba/Kiki Effect - https://news.ycombinator.com/item?id=27885703 - July 2021 (94 comments)

Bouba/kiki effect - https://news.ycombinator.com/item?id=8372550 - Sept 2014 (1 comment)

What consequences are sane for a member who does not pay dues, water bills, etc., for a prolonged period? Turning off the water is illegal here. Liens get ignored. My HOA will avoid foreclosure for years, and has only resorted to it twice in twenty years, but it's the ultimate recourse. Should the other members just suck it up forever?

>Florida ranks 1st of the states on HOA Spy for foreclosure suits brought by associations, with 10,296 of 14,168. The count depends on which courts are indexed in each state as well as on the associations, so it is a count of records, not a rate.

Florida is known for having the strongest "sunshine" laws in the country, that make government records transparent. That makes this statistic pretty useless.

After expressing their utmost sympathies and saying they'd help any way they could, mere weeks later they put a legal "estoppel" on the sale of the house due to some overgrown vines(?), which of course, were never mentioned by anyone.

However saying “all tests have error rates, therefore all tests with error rates are equally bad” is a completely wrong takeaway. One of the things Theranos was doing was using traditional equipment to run the tests as a smokescreen for the fact that they didn’t have any working technology. So the results should have been similar to anyone else running them when that was the case.

But there is a problem that when a company is saying they’re doing one thing or using one technology but your actual tests could be run on unknown combinations of unknown lab hardware depending on the day and the whims of a CEO trying to defraud investors, you don’t know what kind of results you’re going to get.

Real labs have established controls and ship reference ranges for their test results because they control their processes. If Theranos was doing who knows what to keep the test results flowing to keep the illusion up for the primary purpose of satisfying investors, that’s not what you want to see out of your test lab.

> Google tells me the FIT test for cancer has a false negative rate of between 8% and 26%. That appears to be likely worse than the Theranos numbers you cited here.

You’re looking at the wrong thing. The Theranos number above is for tests they determined had to be voided because they weren’t run properly due to procedural errors they identified. Their own incompetence or fraud. And those are just the ones they admitted to!

The false negative rate for the test you Googled is a false negative rate, which is a simple fact that tests can’t reliably diagnose everything even when run properly. Tests look for markers in samples as a proxy for a condition, and those markers aren’t always present at levels that meet the thresholds even if the condition exists. Even when run absolutely perfectly, a test like that will have false negative rates.

The part you’re missing is that the Theranos voided and invalidated tests were due to problems on top of the normal test variation rates because they were from Theranos doing things wrong, which is directly downstream of Holmes manipulating investors.

That's easy enough with a bullshitter in your private life.

A professional bullshitter is a different thing and you might not always have the luxury of just walking away and never encountering that person again (or the problems they create).

We must be living in very different worlds; in my experience these things are never that black&white.

I do agree that it’s very important to know where your boundaries are in life and stand firm on them; and you don’t need to keep people in your life who are a bad influence. It’s just not as easy to make that distinction as some of the comments itt make it out to be …

So pay attention to the sound track and adjust your "suspension of disbelief" accordingly.

You’re on ycombinator, a forum for startups. A non zero amount of people here would do exactly what Holmes did for money.

It did for the author because this is a service she provides. To even suggest that Elizabeth Holmes didn't really harm anyone because she in prison for investor fraud, and then have the nerve to call that distinction "nuance", is absurd.

Theranos generated around 7.8 million test results for Arizona consumers alone. Of those, roughly 834,000 had to be voided or corrected. These were real people getting medical information they relied on. Some received terrifying and incorrect indications of prostate cancer. Pregnant women received results suggesting they were miscarrying when they weren't.

It's like claiming a blood bank supervisor didn't put patients at risk by giving them incompatible blood transfusions, simply because the state only managed to prosecute them for pirating the inventory software they were using.

She did it before for Paul Logan and Anthony Scaramucci, and now is softening the road for Elizabeth Holmes pardon.

She did the same in her newsletter for Logan Paul when he was trying to move beyond his image as a toxic YouTube prankster, and spent months embedded with him and his brother.

Then did the same for Anthony Scaramucci, when he had to rebuild his credibility in the investment world, as his bitcoin hedge fund was about to go nowhere. Pompliano opens with his fascination with astrology. Astrology!

Call monsters what they are. The victims deserve it.

I’ve been around a lot of bullshitters my whole life, and my immediate reaction is to go “this person is full of shit, not worthy of my attention, kthxbye”

But I’ve noticed that most people when faced with the same person can’t help but try and figure out what makes them tick, or their redeeming qualities, or if even anything can be redeemed. It seems to be a social equivalent of “nerd sniping”; most people just have a very hard time declaring “this person is worth 0 additional seconds of my time” and actually following through.

See also ppl in toxic relationships, who will complain to their friends incessantly about how bad the other person is and how they know they need to get out of it, only to rarely get out of it.

There simply is no way to separate what’s reporting and what’s advertising.