hansvm

43 ms·

691.
by hansvm 2y ago
Some comics [0] [1] might be able to get the point across in a way that words seldom do when arguing on the interblags. You're right to call out that we don't know the "ideal" state of anything. I, for one, am excited ab
692.
by hansvm 2y ago
Not really. The unit economics work out heavily in favor of wind even with slower trips and absurd wages and the fact that oil has its externalities pushed to other people. Ignoring local manufacturing minima, the reason we don't do mo
693.
by hansvm 2y ago
It just takes money. $100k-$300k/yr is plenty to have your pick of pretty good people, especially if there are any perks like the food being halfway decent (should be basically a given if you have to pay the chef a lot anyway).
694.
by hansvm 2y ago
The problem is that nefarious actors aren't physically barred from the data. If China, Big Balls, Zuckerberg, or anyone else want to access that data then they can just remove that check. More importantly, the thing you're asking
695.
by hansvm 2y ago
Purity, interestingly, is not a requirement. You can represent any closure as something implicitly captured, and you can represent mutable state as an immutable input with a different, related, immutable output. For any autodiff library not
696.
by hansvm 2y ago
I didn't see a DSL anywhere, just normal JS code. As to what it adds? - It's more accessible to a wider audience (and looks like how you'd implement autodiff in most languages) - It runs in the browser trivially (powering tho
697.
by hansvm 2y ago
That expiration is impossible to enforce. If you have the data and the cert, you can use it whenever you'd like, and the only thing preventing you from doing so is some piece of software voluntarily choosing to comply. What that means
698.
by hansvm 2y ago
Weren't there major problems with the current CVE implementation, especially with the waves of script kiddies and AI tools spamming the database and the fact that projects who take security seriously have little to no say in the "
699.
by hansvm 2y ago
Yeah, the math problem is the other direction -- a thousand things being named after one guy, or a thousand things being named after the simplest English word which could possibly apply.
700.
by hansvm 2y ago
It's a little different. These are systems which are explicitly able to achieve better or worse outcomes by tuning the cost, in ways that aren't especially configurable otherwise. For an HTTP API, you can read the docs and use the
701.
by hansvm 2y ago
No worries. I'll get around to it. I was just curious if you might've explored a bit. Thank you.
702.
by hansvm 2y ago
They're obviously objective. That doesn't make them good. Think of Harvard's goals, maybe some more complicated version of a combination of: 1. Meritocracy: Give a chance to the students with the best innate chance at real-wo
703.
by hansvm 2y ago
Does it suffice to load the content with JS or WASM to keep them out, or are they using some sort of emulated/headless browser? If they're running JS or WASM, can the JS run a few calls likely to break (e.g., something in the WebG
704.
by hansvm 2y ago
The people I've talked to just don't believe it can happen to them. They're going through the normal immigration channels, only getting abortions when medically necessary, and limiting their anti-Trump speech to the few quibb
705.
by hansvm 2y ago
There is a meaningful distinction. They only use backprop one layer at a time, requiring additional space proportional to that layer. Full backprop requires additional space proportional to the whole network. It's also a bit interestin
706.
by hansvm 2y ago
It's a neat idea. It's not too dissimilar in spirit from gradient boosting. The point about credit assignment is crucial, and that's the same reason most architectures and initialization methods are so convoluted nowadays. I
707.
by hansvm 2y ago
The problem is that the 12 month contract is (a) pulled out of their ass and not reflective of the real costs for either party, and (b) until very recently not even disclosed (even in recent months, there are plenty of reports of cancellati
708.
by hansvm 2y ago
GP was right. I was pretty sure your interpretation is correct, but I've seen enough things over the years that I was curious if there were any more details about actual bricking.
709.
by hansvm 2y ago
"bricks"?
710.
by hansvm 2y ago
Why, exactly, isn't that a constructive argument? eastbound> We need Trump because the economy is in the shitter fabian2k> He's actually making it worse eastbound> Well, the other side wouldn't have made it better Ho
711.
by hansvm 2y ago
You're maybe strange (probably not, when restricted to people interested in code), but wrongness hasn't been proven yet. I'd push back, slightly, in that you need to encode those abstract rules _somehow_, and in any modern pa
712.
by hansvm 2y ago
It's a matter of framing. Think of any of the following: - Refreshing daily "points" in some mobile app (handling the clock running backward, network connectivity lapses, ...) - Deciding whether to send an marketing e-mail (h
713.
by hansvm 2y ago
> You can really rely on everyone driving a motor vehicle to obey practically every traffic law every time. That is, they stop for red lights, they stay in lanes, make legal/safe turns; they generally stay near the speed limit and d
714.
by hansvm 2y ago
Oh, sure, the complaint wasn't about SQL per se, but that making complicated things easy to write isn't necessarily a win. Even if a big query perfectly represents the problem, and even if you can use compiled SQL to match hand-wr
715.
by hansvm 2y ago
Some algorithms are much more easily represented as SQL or that plus a little wrapper code. E.g., somebody re-wrote E-Graphs Good using SQL as their primitives, and that viewpoint made a bunch of other features and optimizations obvious. St
716.
by hansvm 2y ago
> focus on [...] the few points that brainpower is really needed The person you're responding to is talking about it from an educational perspective though. If your fundamentals aren't solid, you won't know that exponentia
717.
by hansvm 2y ago
A mathy construction like in the article is probably important for the full conjecture, but isn't this concrete case just an instance of the bin-covering problem? Your discrete items are the log of each prime factor (included according
718.
by hansvm 2y ago
Interesting, that still works? I first heard about that a decade ago.
719.
by hansvm 2y ago
> If they buy your entity (stock transaction) it comes with all the legal liability For dying companies, most of the time it's fraudulent and illegal to create a transaction divesting the good assets from the bad debts. Why is that
720.
by hansvm 2y ago
I voted for not him and persuaded a number of people to do the same (not that it mattered in my state anyway, other than reducing his popular vote margin). I'm not yet sure what more I could have done.

More ›