IJBMer Updates

Stormtroper

But you already learned that years ago.

lrdgck

You got me here.

Stormtroper

So my mother tested positive for COVID-19.
She's been feeling okay-ish so far, but still.

glennmagusharvey

fingers crossed it doesn't get any worse than just a positive test

lrdgck

Watch out, but at least it's a good start.

Stormtroper

Thanks guys.
I think it's been long enough to say she'll be fine.
Also today's my Birthday.

lrdgck

That's odd, I could like swear you had one like a month ago. Time flies these times.

Also, that's good to hear. Keep an eye out for aftereffects, but one can live give or take some wheezing.

glennmagusharvey

Happy Belated Birthday, Stormtroper!

Stormtroper

Thanks guys.
I was feeling sick all day long, I got one of those quick COVID tests but It turned out negative. COVID or not, it hit me pretty hard.

lrdgck

Not Me 1: my neighbour hanged himself. I wonder what does it take to do it.
Me: *mmmffmmphh*
Not Me 2: You were about to say "rope", weren't you.

glennmagusharvey

things i know about Morbius so far, from simply people memeing on it
1. It is a recent movie.
2. It is about at least one vampire. The vampire is probably named Morbius.
3. It is regarded as a bad movie.
4. It is a current meme.
5. Tommy Wiseau did not star in it.
6. There apparently isn't any part of the movie where Morbius's feet are connected to his head with a half twist to make him infinite.

Stormtroper

I was feeling sick all day long, I got one of those quick COVID tests but It turned out negative. COVID or not, it hit me pretty hard.

Apparently I forgot to post about the fact that what I had was real-deal COVID.
The doctor says by Monday I should be fine and dandy. For the most part I feel fine, though I'm still coughing.
On other stuff, I feel like a lot of the common confusion with NP-completeness would go away if NP were called NDP or otherwise it was clear that the N means "non-deterministically", not "not".

lrdgck

So, welcome to the Covid Club?

I was off-computer for like a week, part out of convenience because I had some family business to attend to, but mostly I guess to see if I can. Not offline, though. You know, cell phones these days.

glennmagusharvey

I have yet to actually figure out what the whole "P=NP" thing means. I basically just know it as a math nerd meme.

Meanwhile, I've been rather out-of-it for the past week or so.

Stormtroper

IJBM: "Heat" should mean something like temperature times mass, something more in accordance to its everyday meaning of "quantity of hotness", while the thermodynamics thing could be renamed to, say, "entropic work" instead.

lrdgck

In fluid mechanics, "density" means what a layman may call "weight", and what a layman sees as density, is probably really just viscosity.

And there's a chance what might be called as "viscous" or something is just sticky.

(I mean, I'm speaking from my practice, which is obviously not Anglophonic, but I believe the bulk of the joke holds.)

Stormtroper

On out first day of fluid mechanics class our Prof. made a point out of the difference between density anda viscosity, something I thought was unnecessary.
Similarly, there's the thing where "liquid" and "fluid" are used interchangeably in common parlance, disregarding that gases are fluids too.
Also using "non-Newtonian fluids" to refer to Bingham plastics (which aren't fluids).

glennmagusharvey

> Bingham plastics

holy crap

TIL: the proper technical term for the "phase" of substances like mayonnaise, toothpaste, sour cream, etc.

lrdgck

solid-liquid-gas doesn't even begin

Stormtroper

FWIW I couldn't remember that term off the top of my head and had to look it up.
I see I missed a post:

I have yet to actually figure out what the whole "P=NP" thing means.

It's not hard to understand, but it requires learning a few things in the correct order that might get you lost if you just Google them up:

• Decision problem: a problem where the answer is either yes or no.
• Polynomial time algorithm: an algorithm whose running time's growth with respect to the size of its input in a worst-case scenario is bounded by a polynomial. (Note that in the cause of numerical inputs the size is logarithmic with respect to its numerical value.)
• NP (Non-deterministic polynomial-time) problem: A decision problem where, if the answer is "yes" and you're lucky enough, you can solve it in polynomial time. e.g. the Euler circuit problem: can you traverse a graph by passing through each of its edges only once? If you trace edges haphazardly you may find such a path on your first try, and it'll only take polynomial (lineal) time.*
• P (polynomial-time) problem: A decision problem that can always be solved by an algorithm that runs in polynomial time.

Note that all problems in P are also in NP, for example, the Euler path problem above is in P (if every vertex is connected to an even number of edges (and a few other gotchas don't apply) then yes, otherwise no, and you don't have to rely on luck for this to be doable in polynomial time). If instead of traversing edges once we check for traversing vertices once then what we get instead is the Hamiltonian path problem, you can still get lucky and find a path early on, and there are ways to reliably see whether there's one such path; an obvious way is to try all up to n! combinations of n vertices and you can do It much better than that, but there's no known way to always get a correct answer in polynomial time, it is thus not known to be in P.
NP-Hard: A problem that is at least as hard as all NP problems, that is, if you know an algorithm to solve it you can use that same algorithm to solve any NP problem, thus proving that it can't be any harder.
NP-Complete: A problem that is both NP and NP-Hard.

Now, the easiest way to prove that a problem A is NP-Hard is to prove that another problem B which is known to be NP-Hard can be translated** so as to become an instance of A, thus whatever it is that you can do to solve A you can also use to solve B, reducing B to A (it's unfortunate that "reducing" is what you do when you turn a probably-smaller problem into a bigger problem). I remember linking here a video constructing a boolean (3-SAT) circuit in Super Mario Bros., showing that the game is NP-Hard.
There's an entire network of NP-Complete problems with known ways to reduce them to other NP-Complete problems, if you know an algorithm for one, know there's a way (though not necessarily an obvious one) to use it to solve all the others. For this reason finding a single polynomial time algorithm for one such problem proves that all of them are in P and earns you a million bucks from the Millennium Prize. Despite so many problems known to be in P and others known to be NP-Complete, not one is known to be in both. While we're at it, not one is known to be between both either, though there's a candidate.

Now, this (and running time in general) gets talked about sloppily around the 'net (I assume it's because it's talked about by programmers who don't have to be very into the theoretical side of computer science), so some stuff may cause confusion, it did for me before I started really learning it:

• As mentioned above, the "N" does not mean "not" and all P problems are in NP.
• Special cases of NP-Hard problems aren't necessarily NP-Hard, for example, k-SAT is NP-Complete, because for k >= 3 it is, but (Assuming P != NP) it is not for k=2 and obviously so for k < 2.
• Instances of problems can't be hard. For example, for every graph out there, whether it contains or not an Euler path will be correctly solved either by an algorithm that blindly outputs "yes" or one that blindly outputs "no".
• A problem that is bounded in size by a constant can't be hard by virtue of the fact that an algorithm that looks up a list of all (finite) possible cases for an answer will solve it in constant time.
• The NP thing is about decision problems, it doesn't apply to other problems and this comes up sometimes when it's not obvious what a certificate is or what "getting lucky" means in a non-decision problem, so for example, even if you know the optimal route in the traveling salesman problem AFAIK there's no way to prove that it is optimal short of actually solving the problem, which you aren't going to do in polynomial time. IIRC this also comes up when abstracting problems and treating them as N->{0,1} functions, though I barely got to this part.
• Keep in mind that for a lot of "normal" problems there's usually a decision version of it, for example the decision version of factoring is something like "does n contain a factor between a and b?"

* You may also hear about NP in terms of there being a certificate that can be checked in polynomial time to prove a "yes" answer, e.g. a possible path in the Euler path problem. These two definitions are equivalent, thought I can't remember why "certifiable" implies "luckable". If there's an NP problem where a certificate is something other than a solution to the problem, I've never heard of it.
** In polynomial time, although in my experience this is rarely an issue.

Man, this got much larger than I expected. I'll take this as a sign that deep inside I want to return to studying this stuff. I probably made mistakes there but it's long and it's late so I probably should wrap it up.

Stormtroper

So after a couple months of ignoring the hype, I began watching Arcane. I'm wondering how it can be so good considering the source material.

lrdgck

I've only ever heard good things about it, for sure.

Stormtroper

So I was reading the Wikipedia entry for Ramón Valdés, the actor best known for playing Don Ramón (AKA Ron Damón) in El Chavo del 8.
mfw

Stormtroper

It occurs to me that even for big teams with huge budgets, mascot costumes aren't as well-manufactured as fursuits.

lrdgck

The secret ingredient is love.

glennmagusharvey

ironically true, in a way

also, note to self, re-read Stormtroper's explanation of p = np sometime

Stormtroper

Glenn whoever posted here now was supposed to bring up the Shinzo Abe happenings.
Though, for a relevant reply, I've noticed I'm becoming interested in math again, although I'm finding it hard to dedicate time to pursuits like that these days.

lrdgck

I'm interested in math all the time, except it's the kind of interest where you ask her how she's been and help do her chores while she brings Chad to her apartment.

Stormtroper

The SCP Foundation has been coming up for me a lot recently, and in some important ways, too. Maybe it's no longer an obscure thing? Odd for something I first read about over a year ago.
Also, I wonder if it's a healthy thing for me to keep bringing Up people being wrong on the internet (more math stuff).

lrdgck

I found out about SCP a good while ago, through either TVT or 1d4chan.org (the /tg/ wiki) I think. Always felt like they are trying too hard to be grimdark and edgy. Then I figured people still are into that, and I was like, wait a minute, I dissed it before it was cool, sure I will diss it now that it is.

People wrong on math, I would guess the kind of math an average person is going to be most wrong about is statistics and probabilities, amirite?