It sort of reminds me of when palm-pilots (circa late-90's early 2000's) used short-hand gestures for stylus-writing characters. For a short while people's handwriting on white-boards looked really bizarre. Except now we're talking about using weird language to conserve AI tokens.

Maybe it's better to accept a higher token burn-rate until things get better? I'd rather not get used to AI jive-talk to get stuff done.

One of the companies around today or in the near future will be the one who makes it work at a practical scale. It will have enormous impact, but I think it will be a slow-burn kind of thing as making effective use of quantum computers will take a long time to evolve, IMHO.

Unfortunately, Google and IBM are also working on this stuff and they have deep pockets. They might do it, but even if they don't they may very well decide to acquire whoever does.

These stocks (IONQ, RGTI, QBTS, XNDU) are a sort of thinking-man's LOTTO ticket which will have its numbers called anytime within the next 5 to 20 years (probably closer to 20). I think they're worthwhile to hold in affordable quantities to see what happens. It might hit big, or it might fizzle out for a variety of reasons. There will also be some hype-driven market sugar-rushes along the way that are an opportunity to rake in a modest profit. This has happened already with IONQ, RGTI and QBTS earlier this year. It will certainly happen again when the patagonia-vest people get jazzed about something.

I mean, even Andresson-Horowitz was taking NFT's seriously as though they weren't a scam only a few years ago (https://a16z.com/the-nft-starter-pack-tools-for-anyone-to-an...).

These people are also looking (and funding) quantum computing companies as though quantum computing is right around the corner after AGI.

They need to cool their jets. AI is certainly a worthwhile and super important development, but it's still possible to go overboard with it.

It's actually very pleasing to work with. I wish there was more stuff like this. Lispy programming languages and CAD seems like a natural fit.

That said, python is preferable for most people.

We've long made fun of excel-jockeys getting carried away with VBA, but they came into being because engaging with turgid and expensive software companies to do important but small jobs was such a pain in the ass. This is the start of a new era, and while I am sure we're going to see some wild fiascos, it is a move in the right direction for people that need to solve problems with computers.

They're wise to the fact that "the Stable Genius" isn't going to try anything violent with Denmark/Greenland, but they still want to prevent him thinking about just stealing territory "peacefully."

I find the opposite is true. AI-written copy is an instant turn-off to non-tech folks but many tech-focused people tolerate it.

The contractor (who was certainly incompetent) probably looked at a bunch of nightmarishly complex identity API's and said "F** it!", combine that with being grossly underpaid and you get stuff like this.

It's a bad situation, of course, and involving threatening lawyers makes it even more ugly. But I can understand how a very small business (knowing nothing about IT other that what their incompetent contractor told them) might get really offended and scared shitless by some rando giving them a 30-day deadline, reporting them to authorities, and demanding that they contact all affected customers.

But really, are CEO's the best people to assess productivity? What do they _actually_ use to measure it? Annual reviews? GTFO. Perhaps more importantly, it's not like anything a C-level says can ever be taken at face value when it involved their own business.

I have seen absolutely meticulous lab notebooks before. Each page numbered and dated, cut-outs of graphs taped into the pages, that classy light-green grid-paper. Near flawless penmanship in black ink, with the rare correction crossed out, dated and initialed. Bibliographic references following a strict format in handwriting. Footnotes, FFS.

I've tried, in grad school, 20 years ago to get into the practice. Mine sucked. Non-stop, distracting corrections, maybe a dozen or more per page. Whole swathes of the notebook consisting of deep useless rabbit holes that started with a mis-conception or brain-fart, wasting space, making it a chore to even review what I was doing. I don't think of myself as particularly talented (maybe somewhat better than a fraud). But there are lots of folks like me and much smarter that have the same experience with paper notebooks.

I think really useful notebooks are something that is learned through practice, focus, and mentorship. But there are tools that are much easier to use these days. Notebook-based stuff like jupyter. I like quarto with ipynb myself (though it's not without occasionally infuriating problems).

> Absolutely not.

And what do YOU mean, "absolutely not"? You have no more say in what happens than anyone else unless you're high level politician, who would still be beholden to their constituents anyway.

And yet big science, like particle accelerators, STILL gets funding. There's plenty to go around. Sure, every once in a while a political imperative will "pull the plug" on something deemed wasteful or too expensive and maybe sometimes that's right. But we STILL have particle physics, we STILL send out pure science space missions, there are STILL mathematicians and theorists who are paid for their whole careers to study subject matter that has no remotely practical applications.

Not everything must have a straight-line monetary ROI.

You might be right, but until you get a professional diagnosis you can't really be sure. Hacker-news will disagree but it is impossible to be objective about your own mental health. The good news is that if you do have adult ADHD, it is treatable (much more so than other conditions like depression).

Some people might try to spin it as something cool, but that last "d" stands for disorder. It's a disorder and NOT a "founder thing", regardless of what Paul Graham thinks. ADHD can do enormous damage to your life, relationships, and professional development.

That said, if you don’t like it, disregard it. No one is forcing you. I think it has too much emphasis on protein but that’s just me.

These guidelines theoretically could influence school lunches. Will it make them worse or better or change nothing? ¯\_(ツ)_/¯

One would think there are PCR-based services that do this? That would be the gold standard for this stuff, and it could easily scale enough to become economical, but to my knowledge there are no commercial mold testers that do this.

It wasn’t really a “design flaw” or “weak link” as much as it was management disregarding the warnings of engineering staff. The cold temperature limitation was known in advance by the Morton Thiokol engineers but their management refused to relay the warnings of engineering to NASA and NASA was under pressure to fly. IMHO this was a failure of multiple, mostly organizational, systems rather than “one weak link”.

Did the economists mis-name their own theory?

The github spells it out much better: https://github.com/QwenLM/Qwen3-VL?tab=readme-ov-file#cookbo...

It's a vehicle to sell "access". The greed is only half of it.

The worst part is that they're selling access to foreign interests who pay them off. These people can't exactly show up with bags of gold to bribe King Sh*t Gibbon (yet), crypto is the next best thing.

Many of you may find it shocking or unbelievable, but literacy is slipping in many parts of the US (like Philadelphia). The number of functionally illiterate people is increasing, schools are failing to educate students for a constellation of reasons.

The reality is that we instead suffer from a "tyranny" of illiteracy. I think those folks in their ivory towers, like upenn, should help to address that before starting the pearl-clutching about what has been lost because of widespread literacy.

What is this ultimate precision? I imagine that at some point, even the most modest relative motion at ordinary velocities would introduce measurable time dilation at fine enough clock precision.