makeset

That reminds me, years ago I was on the corporate buyer end of a sizable project, and everything was rolling smoothly toward closing, until a fellow engineer and I went on-site to see the code. It was as if it had been run through a human obfuscator. The entire C codebase was a single main() function in one file, all "subroutines" copy/pasted over and over, cryptic variable names, no comments, no indentation. The "programmers" had to convince us this wasn't a prank before we walked out in shock.

That was especially depressing because the long series of missed deadlines and public demo disasters in the movie (as well as the very buggy eventual release when I later bought the game) gave the impression that his programming skill left a lot to be desired.

Constraint Satisfaction Problems can be modeled as Hopfield-type neural networks, such as Boltzmann machines (RBMs of deep learning are a variant) which can be tuned using simulated annealing.

There was plenty of research on this around 1985-1995 (the "second coming" of neural networks), but it died out with the hype because it was never an actually practical way to solve CSPs. Given the recent innovations in deep learning, it should eventually pick up again, to enable deep learners to solve constraint satisfaction subproblems within the same integrated architecture.

The math in The Elements of Statistical Learning is quite basic, if occasionally tedious. Any college junior in STEM should have taken enough calculus, linear algebra, and probability to work through it.

> they seem to have started offering cheaper upgrades recently.

So much cheaper, in fact, that the ebook+audio bundle often costs less than their monthly Audible subscription which lets you buy one audiobook per month for $15.

Fine, but you being vegetarian wasn't part of my objection. You're the one making assumptions. I don't eat tofu for ethical reasons, and I shouldn't have to justify them over your own to prevent you from violating my bodily integrity by deception if I were to eat at your table. Your argument is still deplorable and morally corrupt.

You happen to be wrong, I am a counterexample (I don't eat soy, and yes, for moral reasons), but I won't explain, because I don't have to. And what I am saying is that I shouldn't have to. Sure, you can decide to believe whatever you want on your own, but when it makes you justify feeding things to me that I have resolved not to eat, that's downright violating.

Do you think people with different sexual preference than you also "should still at least try alternatives when they are presented"? When it comes to something as personal as putting food in their own body for their own nutrition and enjoyment, nobody should do anything just because you think so.

Hell, no.

People may be eating what they choose to eat with just as much moral conviction as your vegetarianism. That is not for you to decide.

You have no right to deceive others about what they are putting into their body, let alone in hopes of "convincing" or converting them.

You have no right to physically force your own morality upon others without their informed consent.

Think about what you have just argued for here.

The only time I write paper checks anymore is for larger charitable contributions, where I assume the organization would prefer the handling burden over processing fees.

Due to the anti-inflammatory effect of low temperature, cold showers soon after resistance exercise may be attenuating muscle hypertrophy response.

Of course, as with most factors other than sleep and nutrition, the magnitude of that effect will likely be unnoticeable to all but elite athletes striving for that 0.5% edge over the competition.

> Add some noise

This actually makes the dataset harder to fit to. It is not the same thing here as the "training with noise" method where random noise would be added to each batch, as an alternative means of Tikhonov regularization.

That's the standard recommendation from sports nutrition texts, and I've certainly found it to be very true for myself. Over many cutting cycles as a strength athlete over the years, I always lost high ratios of muscle:fat during IF-style cuts, and augmenting with protein intake during long breaks between meals has made a tremendous difference despite my advancing age.

The total size of the HFT pie is indeed minuscule, considering the amount of hype it gets.

Past the high infrastructural barrier to entry, and assuming they can stay ahead of the competition, the consistency of returns does seem lucrative, but the actual magnitudes of the opportunities are hard-limited by the nature of the time scales. You can't move any significant amount of capital in milliseconds and expect to profit. Compared to the average hedge fund, they are playing with pennies.

Do what you will, but that was not a personal attack.

Yes, I'm well aware of R's many faults, I have my own long list of R caveats I hand to new hires, but not bothering to learn the damn language is no reason to complain about it. First, RTFM.

R's server sockets can't select(), so unless you've reimplemented that as well, don't count on handling concurrent requests with that web server.

That argument has been made countless times in history, with just as much conviction, but replacing human/animal with black/white, male/female, heterosexual/gay, Christian/other, etc.

> I don't see how it's anyone else's business what I do with an animal that belongs to me.

This is why we have animal welfare laws, because there are people like you out there.

It makes me sad to see this comment here. This is precisely how pseudoscience sells and proliferates: subjective attribution of all success to the hyped bunk.

Quirk diets work because adherents accidentally get the calorie balance in the right direction by limiting/changing dietary habits. It's not the quirk, which might maybe contribute a tiny effect if it's not entirely bogus in principle. Yet people hail, "The diet works! The guru was right! All hail the Guru!"

Ask and you'll hear the exact same refrain for homeopathy, acupuncture, rhino horn powder, shark fin soup, albino body parts, ...

I've cut out all sources of caffeine completely, except for a glass of tea right after I wake up where I pour away the first steep.

I use NRR 33 earplugs and at least one white noise machine going, sometimes two, to avoid any noise pulling me back from near-sleep. The room is dark, but I wear a sleep mask just in case.

I'm averaging 1-2 hours with those. I'm doing the sleep hygiene thing, where I don't do anything in bed except (try to) sleep and sex. I think you're also supposed to just get up if you can't sleep, but if I do that, I just never sleep. I come back even more exhausted, and I'm still looking at the same damn wait.

Mental activity can definitely make it worse, but I haven't found a meditation/visualization sort of trick that really helped. The best I can do is psyche away the frustration and just... wait.

Alcohol/pot help maybe a little, but then I don't sleep deeply at all, always wake up really tired. I haven't tried prescription drugs. I'm told they have the same effect on sleep quality, and I don't want that. I do get much-needed good sleep once I'm there. It just takes too damn long.

I'd love to hear if anything's worked for you.

Sounds cool, but falling asleep takes me 1-2 hours on a good day. If I had to go through that any more often than once a day, I might stab myself.

[1] is also here:

https://learnxinyminutes.com/docs/lua/

No, I meant that he was squinting. What is condescending is you making assumptions about whom I don't know and what I'm unfamiliar with.

Like many modern languages, R supports multiple programming paradigms, including sequential, object-oriented, and functional, but it should take more than a couple of functional features to declare something a full-blown FPL. A duck-billed platypus is not a duck.

Yes, you can write R code in a functional style in places when convenient, but one floor down it's still running that loop you didn't want to see. It's a language based on mutable values and sequential execution and looping, not pure function calls and recursion. The syntax itself is very much sequential, unlike stereotypical FPLs. Even the lexical scoping design relies fundamentally on mutable environments. So yes, people do seem to enjoy proudly declaring R a FPL citing a few features, but as I said, those features make R as much a FPL as Python.

Yeah, lots of squinting needed, or a terribly shallow understanding of FP to say that. Merely providing first-class functions and map/reduce idioms does not a functional language make. Python has those, so is it also FP really? R is as much a FP language as Haskell or OCaml are imperative "at heart."

I also started Songwriting and Jazz Improvisation, and while both had outstanding content and instruction, the peer-grading of assignments was a complete failure and forced me to quit the coursework timeline. Aside from reviewing others' work being a huge time sink, all feedback I ever got myself was completely useless, like "You did everything in the instructions, but I don't like this" or "Gee, I don't know enough to tell you how to do this any better." I would have gladly paid to get qualified feedback.

I knew I'd seen that name before.

http://bellard.org/jslinux/

> you don't/can't "isolate" development of a muscle

This is not true at all. You may be confusing it with the assertion that you can't isolate individual heads of a particular muscle.

Yes, training has some systemic effect through hormonal activity, but muscle anabolism is very highly specific to local stimulus. Otherwise we'd all be just doing curls and watch our legs grow.

For a comically extreme example of localized functional muscle hypertrophy, search for the armwrestler Matthias "Hellboy" Schlitte :)

I don't know if there was an undocumented way, but the extension explicitly stated you had to uninstall it to stop the timer or to change any settings.