zopa

I'm not a shadcn user, and so as with any project I'm not familiar with, I'm looking to see if if it's interesting to me. If the post is thoughtful and clear and I like the sense I get of their perspective on programming, then great, tell me more. No guarantees but maybe the writer is someone whose software I'd like to use. Claude doesn't tell me anything except that the writer used Claude.

It's the same as with too marketing-speak, which conceivably this is. Maybe the actual work is good but Sturgeon's law, it's probably crud. If I really needed a UI library or whatever right now then maybe I'd dig deeper but in casual browsing HN mode? No time, catch them later.

So maybe all your maybes but who cares? It's not AI that made me think badly of them: I think badly of all software by default and it takes more than Claude to change my mind.

Usually there's one or two job responsibilities among many, that you can do, but not the way everyone else does them. The ADA requires employers to make reasonable accommodations, and some employers don't want to.

So less, the job requires you to stand all day, and more, once a week or so they ask you make a binder of materials, and the hole puncher they want you to use dislocates your hands (true story). Or, it's a desk job, but you can't get from your desk to the bathroom in your wheelchair unless they widen the aisles between desks (hypothetical).

Nit, but if you're thinking of Haskell's text library, it's utf8 now: https://github.com/haskell/text/pull/365

If the goal to drive the yield on T-notes down, then cut spending and raise taxes. Get the budget into something like balance and the Fed will cut rates, and borrowing costs will fall. You don't need some 3D chess with a manufactured financial crisis.

When your rational explanation involves people shooting themselves in the head to cure a headache, something's off. There's nothing to explain here besides a weak and incompetent leader with a deep need for attention, surrounded by lots of enablers.

And it’s a healthy wake-up call when someone who knows close to nothing about our fancy stuff designs a system that we would unlikely think of.

I can't tell if the author asked Kasia how she came up with the idea? For all we know she's doing an MBA in the evenings and just wrote a paper on the history of just-in-time manufacturing in the Japanese auto industry.

Not sure why you’re fighting this so hard, but: GW, Triboro, Whitestone, Throgs Neck, Goethals, Outerbridge, Holland Tunnel, Lincoln Tunnel. Only the last two land in the toll zone. So yes, “most,” and it’s not especially close.

I’m not counting the Brooklyn Bridge, etc, because I was replying to your “ways to enter the city.” The topic as I understood it was “will you pay this toll if your final destination isn’t Manhattan,” and the answer is “Not unless you’re coming from certain parts of New Jersey, and even then you’ve got choices.”

You wouldn’t want it in the CI pipeline, because any model clever enough to find real issues is also going to find plenty of false positives. That seems like too much friction for most open source projects.

I’m not one of the downvoters, but you’ve linked to a list of forty or fifty different projects, many of which don’t seem relevant to this use-case. It’s not too surprising people have nothing to say besides “ugh, more AI hype.”

It’s Manhattan below 60th, not all of Manhattan, so maybe half the island, and it doesn’t include the FDR. Most of the ways to get to Queens, Brooklyn or Staten Island by car won’t be affected — same for the Bronx obviously.

Nah, good AI would run in the compiler and optimize the recursion into something fast.

Have a goal, and measure your progress towards that goal. If "get stronger" is enough, great; but if not, find some challenging activity you enjoy and want to improve at, and do the resistance training that will help you get better. Personally I find that much more motivating; just knowing there are health benefits never really did it for me.

Really really important to write down what you do. Progress is not always fast, but if you keep records, you'll see it, and that really helps with motivation.

IMO podcasts and such are not so good for strength training specifically, though for endurance work they're fine. You really want to give full intensity to getting through that last rep (with good form). You're training your ability to push through discomfort as much as you are your muscle's theoretical inherent strength.

Counterpoint from someone better informed about bipolar would be interesting and valuable. That said, it’s a discussion of an article asking “why is this guy like this?” One potential partial explanation is a mental health condition. Certainly it crosses my mind any time there’s a consistent and self-destructive pattern of behavior that it’s not clear the person involved can control.

I don’t know what’s served by pretending otherwise. If mental health challenges are real and important, and they are, then they’ll have observable effects outside a doctor’s office, and we should be able to talk about that.

Then I’m not sure why you’re arguing that $50 billion would make the problem go away. Care to explain the good faith reason if you have one?

Good news: we’re spending more than five times that on domestic semiconductor manufacturing, already, starting two years ago. https://en.m.wikipedia.org/wiki/CHIPS_and_Science_Act

I’m not sure Apple’s the best example for you. https://www.nytimes.com/2024/01/05/technology/antitrust-appl...

That’s the difference between a plain old tax credit and a “refundable” tax credit.

Can't speak for what the person you're replying to meant (lol), but no, of course it's not nearly that stark. There's a continuum. Concrete everyday topic that comes up frequently? Everyone's going to know what they mean by what they say there. But we've all been in business meetings where people were just putting words together in a plausible way; we've probably all been the people speaking at those meetings.

The fact that it's possible to say things that make no sense, without knowing that they make no sense, proves the point. You or I can come up with a phrase like "the barber of Seville shaves all the people who do not shave themselves," and we can come up with paradoxical phrases without realizing that they are nonsensical. Neither of those would be possible if there were always a strict, one-to-one relationship between speech utterances and facts about the world.

Sometimes we have a specific idea of what we mean by what we say, sometimes we're just putting words together, often it's somewhere in between.

I don't know what the response time usually is in urgent cases, do you? There's a quote on pg 16 of the study that makes same-day sound unremarkable. It starts:

A lot of times the people who have their weapons seized are not having a bad life—they’re having a bad moment.

It's an ex-prosecutor's illustrative hypothetical, so take it for what it's worth. Still the same logic applies if we're talking about bad weeks instead of bad evenings.

Thanks for the link to the paper. But I don't think that 10-20 seizures per prevented suicide estimate you mention can tell us anything about probable cause or the lack thereof. It's a guess at the rate of prevented deaths from suicide attempts, when the subject used a something other than a gun, but would have used a gun if the cops hadn't taken it from them. It's not an estimate of how many suicide attempts were prevented: it's just a measure of how much less deadly the suicide attempts that happened anyway were. A gun seizure is a big dramatic intervention; it's pretty plausible that it sometimes interrupts a bad moment that's essentially a one-off, and sometimes leads to people getting the help they need. Neither of those effects play into the estimate you cited.

Just on the legal standard: it's probable cause that there's a risk, not probable cause that the risky event will happen. If you have 51% certainty that the subject has a 25% chance of harming themselves or someone else, that is likely to be good enough. And while it gives me no joy to defend the US criminal justice system, to me that's appropriate here. Living with someone who's making violent threats towards you or themselves is no fun at all, even if there's only a 1 in 4 chance that they'll follow through.

Nowhere anywhere ever has perfectly and strictly enforced all its laws, or even come particularly close. Prosecutorial discretion isn't something the US invented.

I use nix-shell, and mostly I love it. But it’s important to be aware that the above means “Go get the latest(*) versions of python, pillow and ansicolor and run this code in an environment where they’re available.” It doesn’t do any version-pinning of your dependencies. That might be what you want, but maybe not: it’s frustrating when a script that worked yesterday won’t work today, or will only work after some big download.

My own rule of thumb is that nix-shell is great for quick one-offs and for sharing environments. For local tools and anything else I’m sharing with my future self, it’s usually better to write a nix expression and install it, which gives me access to Nix’s (excellent) rollback system, and lets me upgrade on my schedule, not upstream’s.

* - ‘Latest’ according to whatever Nix channel checkout currently applies. Which you can change, of course, but the point is it’s external to the script.

It’s a great choice for the JS ecosystem for the same reason it’s a terrible choice for the JS ecosystem: JS dependencies are a lot, and they sometimes want to do strange things at install-time that Nix frowns upon. There’s definitely an upfront cost, and a maintenance burden as well. But the flexibility and the control over what code you’re actually running could still be worth it.

“Thanks for coming to us with this, we’re looking at it right away” wouldn’t take a lot of time or commit then to anything

You can also do Ctrl-Z to pause the running process, and then `%1; say done` (or whatever) to restart the first queued job and then run the new command. Avoids the interactive issue

Taking something from a research prototype proof-of-concept to production absolutely counts as innovation. In some ways it's the most difficult and impactful phase of innovation.

Likewise there's an early DARPA project for basically any technology you care to name. That says a lot about US budgetary politics and not much about the tech itself. Spend a few hundred million on most research and development and it's a big-government boondoggle; make it a national security appropriation and no one in Congress will bat an eye.

Those young engineers have worked really hard and done some extremely impressive things. Perfectly possible to acknowledge that, whatever you think of the majority shareholder.

I hope I wasn't unclear, I'm not stuck on testing per se. I want Amazon to stand behind what's sold on its site. I don't really care what goes into that: if it needs to physically test products to do that, then it should, but if there's a different/better/cheaper way of getting the confidence to take responsibility than that's fine too.

No argument that that's burdensome and makes it more expensive to run an online marketplace; of course it is. But what's the alternative? Pure buyer beware doesn't work when I can't physically inspect what I'm buying and there's often no way to know much of anything about the seller. Hassle-free returns are nice when you're only out money, but undisclosed drugs in supplements can be genuinely dangerous. A refund isn't going to fix things in the really bad cases.

This is how liability has worked in the American and British legal systems for at least three centuries: when there's been harm done, the person harmed can recover the full value of the damage from any party that helped cause the harm. Plaintiffs don't need to worry about who was most at fault; that's for the defendant to work out with the other parties afterwards. Similarly the FDA can't and shouldn't need to start a separate regulatory action for every seller on Amazon's platform. It's totally appropriate for them to lean on the large domestic company with the fixed address and known officers, and rely on Amazon to enforce compliance with the law down at the seller level.

Griffin is pulling a bit of a fast one though: comparing Apollo to what the Space Shuttle turned out to be, rather what the Shuttle was promised to become. If Congress and the country had known that the choice was between a continued Apollo program and the Shuttle we actually got, it’s all too likely they would have chosen neither. Apollo cost too much to sustain; it’s no defense that the shuttle turned out to cost even more.

There’s a lot of hindsight bias here. We know now that the Shuttle was a bad design that was never going to give us cheap and routine access to space. But I don’t know that it’s fair to expect that to have been clear at the time, before it’d been tried.

I think you might be underestimating the value-add---at least based on the existence of this thread! Yes quality copies are out there, but easy to find for the Editor-in-Chief of Standard Ebooks doesn't mean easy for everyone. I suspect plenty of people would find a trusted, no hassle source for a quality print copy worthwhile, just for the simplicity and convenience. Though I totally respect not wanting to waste ink and kill trees reprinting something that's already widely available.

It's easy to say "don't allow this to happen"; I'm curious about what that actually translates to in practice.

It translates into "don't allow this to happen," because any other standard can and will be gamed. So for instance testing a bottle or two at random would work fine if Amazon really does test a random and representative sample of what's being sold, and to me that's a reasonable level of diligence to expect. But if Amazon Testing emails the supplier: "Please send over a batch of X MAX SUPER ENERGY so we can test if for these substances which we hope we won't find, and make absolutely sure what you send is the same thing you're selling!" -- then that will work somewhat less well. There are endless ways for Amazon and sellers to wink and nod and skirt the intent of the rules, if Amazon is just checking a box for the FDA, and doesn't actually care.

The way to get Amazon to care is to hold it responsible for the outcome. Of course there will be sellers that find a way to skirt whatever process Amazon puts in place, but that needs to be Amazon's problem. Amazon can survive taking its lumps when it messes up.

This is where we need to decide if we're doing physics or philosophy. Physics is about answering specific, fairly practical questions, and the questions tell you what parts of reality you can afford to ignore. If you want to get at the true nature of things in a metaphysically satisfying sense, physics will always disappoint you.

Is it part of the essential nature of a building to collapse? Sure, I suppose, as with all things. But I'm only going to be in this this coffee shop for another hour: to me, for my purposes, it's stable. If instead I were buying the building I'd want a much more detailed model that considered termites and the risk of earthquakes and all sorts of things, but I still wouldn't care about the date of the next ice age that will scour the landscape clean.

One thing to keep in mind is that our mathematical methods might as well be approximate, because we our measurements always will be. Precise calculations on fuzzy data are a waste of time. Not that there's anything wrong with philosophy! But it's not super relevant to understanding where harmonic oscillators are a useful model.

It's quite absurd to assert that ELIZA from 1966 outperformed GPT-3.5.

They don't. The researchers ran a modern ELIZA implementation alongside the GPT models. They're not citing ELIZA results from 1966.