Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[methylatedspirit]

I don't mind that you talk about it, and I don't really believe in a "too much" of talking about emotions and feelings. We all need an outlet for our feelings. At least it shows you're able to say it. Lots of people just keep their problems pent-up inside them, never saying a word, and that's worse than talking about it.

That's, like, classical toxic masculinity, though I suspect there are analogues across other toxic gender stereotypes.


I originally came here to say one thing: The music I listen to has little dynamic range. I think I'd be right in saying that the most dynamic range I ever regularly experienced was Tame Impala's One More Hour, and that's only -10 dBFS RMS, a DR of 6/14. My music is loud as shit, in other words. I think it's because I listen to pop and a lot of vaporwave/future funk (which is itself derived mostly from pop), and those genres are not at all known for anything resembling dynamic range.

[methylatedspirit]

I've had this analogy in my head for cooperative multitasking for days now. I don't think it's very accurate. I must've gotten it from somewhere, but I'm not sure where.

Suppose there was a teacher (the kernel) and a bunch of kids (processes) in a classroom (OS) together. The kids need to do some kind of large overarching task (running the OS), but they only have a single sheet of paper (the hardware) to do it on.

The teacher hands the paper to the first kid who asks for it. This is done under the assumption that the kid will just pass it back when they're done. Once the kid hands it back, the teacher looks for the second kid who asked. Then once Kid #2's done, the teacher hands it to Kid #3. And so on.

The glaring problem with this kind of multitasking is that a kid might not give the paper back for whatever reason. Maybe they're being a little shit, maybe they're stuck, maybe they're waiting for someone and they're stuck, who knows? And if that happens (which it will), the whole task is basically screwed. Only thing you can do is to end the session, and hope the paper isn't too badly damaged to continue.

[ArchimedesWojak]

Goddammit imgur wont load my honey badger picture

[Yellow Pixel]

Shameful! The honey badger is one of the coolest mustelids!

[dragdeler]

-

[methylatedspirit]

"Back in my day, you could push sound through a real video codec like Cinepak, and you'd get good results. Now, you got your aitch-two-six-five, your vee-pee-nine, and your ay-vee-one, and they sound like robots screwing. *sips Mountain Dew*"

-- Me, after a previous version of this post stated that older video codecs do better on audio compression.

[Iduno]

I know the fact that cats like paperboxes is also very much about it conserving heat, and protecting from sight. But maybe they like paperboxes THAT much because we often keep kitties in them? Some early childhood memories about feeling safe so to say.

I've heard it's an agorophobia thing, which is also why they like hiding in drawers, under beds, behind couches, etc.

[methylatedspirit]

After some thought, I realize that this function I adapted from the u-law algorithm for a forum game



as a sorta-sigmoid (but not really) is a "polarizer" (surely there's an official word for this). It takes deviations from the center of the distribution, and pushes them toward the nearest edge. If I just generate a metric fuckton (which in this case means 100000) of uniformly-distributed numbers in the range [-1,1], and define x₁ = 0, s = 1, y_s = 1, o = 0, u = 255, then put said fuckton into f(x), I get this histogram:



Which certainly looks like the inverse of a normal distribution and their ilk, at least in general shape and "feel". That one collects values near the center. This one collects values away from the center and pushes them to the edges.

[taat]

It's very similar to tanh in that way

[Frumple]

I know the fact that cats like paperboxes is also very much about it conserving heat, and protecting from sight. But maybe they like paperboxes THAT much because we often keep kitties in them? Some early childhood memories about feeling safe so to say.

I've heard it's an agorophobia thing, which is also why they like hiding in drawers, under beds, behind couches, etc.

Cats like boxes so much, you don't even need actual boxes, just the right sort of tape put on the floor in a box shape. We don't really seem to know the exact why of it, yet, but it's definitely a thing.

[Maximum Spin]

Cats like boxes so much, you don't even need actual boxes, just the right sort of tape put on the floor in a box shape. We don't really seem to know the exact why of it, yet, but it's definitely a thing.

That's not really what that article is saying. Did you read the linked study? It wasn't even testing whether cats prefer rectangular delineated spaces to some alternative, and only a minority of test subjects settled in a delineated space at all; certainly they don't seem to have behaved like they thought it was equivalent to a cardboard box. The article is discussing whether cats' visual systems are susceptible to a particular illusion in the edge-finding process as humans' are, which seems to be the case.

[methylatedspirit]

It's very similar to tanh in that way

I suppose it's common to all sigmoid functions, no? They all look vaguely like that f(x), so they'll all do something like that.

Though I'll clarify that the purpose of sigmoid functions is to do a one-to-one mapping of the range [-∞, ∞] to [-1, 1], or at least they make it really convenient to do so. The original purpose of that f(x) is to map the range [-1, 1] to [-1, 1] {ensuring that the function exists at the points (-1, -1) and (1, 1)}, but non-linearly. There's uses for that in telephony and such; see u-law and A-law.

[methylatedspirit]

I swear, calculus (and math, I guess) is just a string-rewriting system. Match some pattern, replace with something else. Match another pattern, replace that with something else. Match yet another pattern, replace that with something else. Repeat until you see the final answer (i.e when the derivative/integral operators disappear).

[Rolan7]

Babies have no concept of exponential growth.  It has to be taught - one of the core lessons, perhaps.  Something can increase, yes - but faster!  faster with more time!
"For how long?" "Faster the more long"
"But how fast??" "Faster the more long"

[methylatedspirit]

Babies have no concept of exponential growth.  It has to be taught - one of the core lessons, perhaps.  Something can go up, yes - but faster!  faster with more time!
"For how long?" "Faster the more long"
"But how fast??" "Faster the more long"

That's strange; isn't human perception more-or-less logarithmic (see Weber–Fechner law)? I suppose that would make exponential growth look linear to babies, since logs and exponentials cancel each other out. They don't have any understanding of linear scales.

That reminds me: I heard that young children and those who have no formal math education will draw an exponential number line if you just asked them to label evenly-spaced marks on a blank number line. The first mark might be 1, the second 2, then 3, 5, 7, 11, 17, 25, 38 (...). It's only when they have some math education, they start drawing a linear number line (1, 2, 3, 4...). I so want the source on that, since I must've heard it from somewhere.