bermanoid

Game engines are different. It's a tiny market relative to web, and really the only people in the market for an engine other than Unity or Unreal are indies, who have near zero revenue and even less willingness to spend it. Those two are so far ahead in terms of features, support, and battle-hardening that you'd pretty much have to be insane to pick anything else if you had paying users.

Always be wary of any market where someone's willingness to try your product is in itself a negative indicator of ability to pay. Hit-driven markets that attract large numbers of non-serious dabblers are extremely difficult to sell tools profitably to, but it's easy enough to get minor attention that makes you think you might have something worthwhile (music production is another one that scatters corpses all over the place despite seeming large at first glance).

That's a great ideal, and I understand the arguments in favor, but the reality is that sometimes in order to test all of the internal code paths, you have to go to extremes when only interacting via the public interface. If I'd have to write 50 lines of extra testing code (or worse, extend various classes to add fake hooks into external dependencies, etc., which is where testing tends to really get messy) to validate that some edge case is handled appropriately, it's sometimes worth skipping that and fiddling some internal state to jump straight to the edge case.

Never fund yourself what someone with 100x the money will fund for you.

If you can point to something about the situation that has changed since the shelter-in-place was initiated that would indicate it's not needed anymore, I'd love to hear it. The virus hasn't gone away, we don't have any medications to make it reliably recoverable, we don't have test + trace ability in place (anecdotally, even if you're symptomatic but not in seriously bad condition it's still very difficult to get tested, let alone if you're not), we don't have herd immunity, there is no approved vaccine, etc. If the Bay Area opens up completely, exponential spread kicks in instantly with a high replication factor and in a month it'll look just like New York.

We are not through this in any way, shape, or form, even if it sucks. We've bought some time. If we did reopen now we'd probably be better off than if we hadn't shut down at all, but it'll still rip through the whole population, and unless we're ready to accept that, we've got to keep measures in place.

Totally agree with pretty much all of this; I'm just arguing that "don't subsidize, ever" is a very extreme vision of capitalism, and doing a one year or less intervention is nowhere close to throwing away the whole system (which GP is implying very clearly elsewhere in the thread, and to be fair I probably should have responded to those specific comments instead of this one).

I agree that when and how we intervene is important to discuss, it's just important to argue the details, not "but muh capitalism" as a way to argue against every government action ever.

I'm 100% in favor of capitalism, but that's not what you're standing up for. Capitalism doesn't mean you never intervene when disaster strikes, it just means that markets are the engine that drive the economy and you mostly let them function unimpeded. When something threatens the proper functioning of the market itself, a functional capitalist system should intervene (in non-market-based ways if necessary) to keep the flywheel spinning when the crisis passes.

Put another way, laissez faire capitalism might be fine when the system is operating in a normal regime, but most capitalists don't take that as a prohibition on taking action when things like wars or natural disasters strike. All of the usual arguments in favor of letting the market just do its thing break down when you're temporarily in an environment that differs hugely from business as usual, and letting the market decide for itself could end up optimizing for the wrong behavior. Specifically, we probably don't want to allow a natural selection event that cuts so deep that anyone without a 12 month cash buffer goes bust, since if every company operated to maintain that it would not be optimal behavior during the steady state.

This looks like an actual log-log plot, but of new cases vs existing cases. Not a super useful view except to point out that yes, growth rate tends to be proportional to infection count (pure exponential growth) until something majorly changes.

A linear plot would tell exactly the same story, just be compressed too far towards the origin to be very readable.

So do most of the major cities in the world. That's a hell of a confounding variable.

Not GP, but also a test prep teacher in a previous life. I always scored well on the SAT and test well in general, so I can't speak to personal improvement there, but a few points:

First, the test prep companies heavily encourage (and may even require?) students to take the test multiple times. That change alone will boost most people's top score by a solid margin, because it's a noisy test. The first response they always gave to people complaining about lack of improvement was "take the test again, and if you still haven't improved, take the course again for free, and then take the test again".

Second, psychometric expertise is great, but the goal of the SAT is not to be impossible to train for except as a secondary thing. It's really hard to do, especially when you are such a high value optimization target and have to build a test that doesn't rely on much specific knowledge and can be quickly scored. A lot of what the SAT courses do is just teach students to make slightly more accurate guesses on multiple-choice questions where someone unfamiliar with test strategy would leave things blank. That alone tends to boost scores, and some of the other strategies are fairly clever to help avoid common mistakes.

Last, though I didn't have room to improve on the SAT, I also taught GRE classes and can speak to my improvement there. The math section is trivial to get an 800 on (it's easier than the SAT math, or at least it was when I taught 15+ years ago), but the verbal section is quite tough if you haven't studied, and in a lot of ways is a glorified vocabulary test, and the reading comprehension sections can be pretty tough as well. Before I trained to teach the courses, my verbal score was a 490 (on the real test, not practice), so I was considering not even trying to teach (there's some threshold you have to hit, maybe 700 at the time?), but the trainer encouraged all of us to try anyways because he said the content made such a difference. After just two weekends of intensive teacher training, I tested again and ended up with a 740. After teaching the course around a dozen times, I'm pretty confident I could have hit 800 without any difficulty, you basically just have to get used to the types of questions that they ask and get in the headspace of the question authors. Just one data point, but I definitely believe that this stuff is effective.

That's a fair criticism of the general SAT - I used to teach those classes, and the test is very gameable in the sense that there really are "A FEW SIMPLE TRICKS!" to learn that have nothing to do with actually knowing the material in a useful way.

But at least some of the subject SATs are not like that at all, specifically the math/science ones. There really aren't many tricks or traps, they really are like normal school tests (if multiple choice) where doing well on them requires you to know the material they cover. Nobody who is "good at tests" is going to 800 the physics one without knowing physics well enough that they'd do well in a freshman mechanics course, and someone who gets a 400 either slept through class or is going to struggle at college level.

Acceleration does not equal velocity squared...

Claiming that some mysterious and hard to define property that we can't measure even in principle "exists" in some meaningful way strikes me as the stronger claim than the skeptical take does.

Why do you think the burden of proof should be inverted? The mere fact that most humans intuitively feel "something" doesn't count for much of anything, especially once you stipulate that p-zombies would vote the same way.

He's also famous for the Chinese Box thought experiment, widely derided by everyone apart from his own students as the most high profile, idiotic, uninformative, trivially debunked thought experiment of all time, which teaches us negative information (in that it actually wastes time bringing up useless shit that otherwise wouldn't receive scholarly discussion except that he's an old white guy that was in the field early).

Searle is an absolute waste, nobody should engage with his drivel, ever, period.

And I imagine many more machine learning tools will take the same name in the years to come, since it's about the most obvious one you could think of other than "brain".

Whatever is popular will survive...

ImageNet Roulette deliberately uses a terrible categorization scheme that has long been acknowledged as so poor as to not admit meaningful results in order to make the highly political point that ML should never be applied to people. There's a reason most people scrub that whole piece of the taxonomy before training.

Good Resnet models trained on ImageNet (the good parts, not just people) tend to result in state of the art results for almost every transfer-learning domain they're applied to.