The tournament in question is 13 days long, the winner makes $100k, and the top ten finishers all make over $10k.

The online tournament a few days later that Niemann was subsequently kicked out of has a $200k top prize, and the top 32 finishers all make over $10k.

And there are lots more tournaments like this.

Magnus Carlsen has an estimated net worth of $50 million. So you can't exactly make subvert-democracy money, but you can make movie-star or pro-athlete money playing chess.

How is that in any way a parsimonious explanation?

It doesn't explain the fact that Carlsen has never made accusations like this over the course of his 15 year career, much of it spent as the top player in the world.

It doesn't explain the fact that Niemann has repeatedly been caught cheating, has confessed to only what he's been caught doing and claimed he didn't do anything else, and then has confessed to having done more when he was caught doing more.

It doesn't explain the fact that Niemann trained with an expert on chess cheating who's been banned from events with cash prizes after confessing to cheating.

It doesn't explain the fact that Niemann played games against other players in the same tournament where there were clear hallmarks of cheating, and that he couldn't explain anything he'd done after the fact.

> They aren't accusing him of running a computer program on another device to feed him moves to play in the on-line game?

That's precisely what they're accusing him of doing.

The thing is, if you run the computer program and make all the moves it tells you to, it's *very* easy for someone to catch that by analyzing the moves after the fact. So, instead, you want to have the computer program running, but you don't copy all its moves. You cheat on one or two moves where it matters.

If your calculations are processor-dependent, that's a bad sign for your code. If your results really depend on things that can be altered by the specific floating-point implementation, you need to write code that's robust to changes in the way floating-point arithmetic is done, generally by tracking the uncertainty associated with each number in your calculation. (Obviously you don't need real-time performance since you're using cloud computing in the first place.) I'm not an expert on Mathematica, but it probably has such things built in if you go through the documentation, since Mathematica notebooks are supposed to exhibit reproduceable behavior on different machines. (Which is not to say that no matter what you write it's automatically going to be reproduceable.

Archiving hardware to get consistent results is mainly used when there are legal issues and some lawyer can jump in and say, "A-ha! This bit here is different, and therefore there's some kind of fraud going on!"

Yep. Generally what happens when there are multiple defendants and no solid case is that the suspects are threatened with a Russian Roulette-style court case, causing one to decide to confess and implicate the other defendants in exchange for a lighter penalty.

I dunno what you mean by "UC isn't even all that prestigious," but the Times Higher Education Supplement rated UC Berkeley as the #8 university in the world and #6 in the U.S., behind Caltech, Stanford, Harvard, MIT, and Princeton. In particular, they feel that it's better than the entire university system of every country that isn't the U.S. or the U.K. I'd say that's pretty good. UCLA is world-class as well, and UCSF is one of the top three or four medical schools in the world. Even the lesser-known campuses like Riverside and Irvine have strong reputations.

Personally, I find it fascinating that if you just randomly put forward and back slashes on a screen then it forms a pattern like that. To me, it's much more interesting than an algorithm that generates mazes. I'd have expected that you just got a bunch of diamonds and things.

In elementary school in the late 1980's, our school librarian of all people -- a nice, fun older lady -- taught us LOGO programming. The class went relatively deep given that the students were all eight or nine years old. Our last assignment was to write a function that would draw a regular n-gon (taking n as a parameter), then incorporate that into a recursive function that would draw arbitrarily deep spirograph-type shapes using a callback function. Pretty much everyone figured it out on his or her own, as I recall. Our "computer lab" at the time consisted of someone going and setting up folding tables in a hallway or the cafeteria and then lugging a bunch of Apple //e machines out of a closet, then tearing the whole setup down after a couple of hours.

In middle school, we had a short unit on BASIC programming, by now on the Apple IIGS. By this time it was the late 1990's, and I'd started teaching myself QBASIC on our home PC; the computers we were using in school were around seven years old by this point.

In high school, I took a semester of computer science as a freshman and a year of "AP Computer Science" as a sophomore. This was largely just indoctrination into OOP. The entire course consisted of writing completely trivial C++ programs which would consist of several objects, none of whose member functions exceeded one or two lines. Nobody really enjoyed the course or learned much of anything, but we were pretty much bound to the AP curriculum so there wasn't much that could be done.

Had I my druthers, I'd design a computer science program for schoolkids by focusing more on the sort of stuff I did as a little kid, which was really conceptually much deeper and certainly a lot more fun.

You *definitely* don't need a PhD to work as a mathematician at the NSA, nor do you need a specific background in cryptography. If you failed abstract algebra in college you're probably not getting a job there, but my understanding is that it's not that difficult to get in so long as you're qualified, a citizen, and can pass a background check. The NSA is the single largest employer of mathematicians in the world -- they're certainly not just hiring the extreme elites.

Did you read his twitter post? It was an obvious joke. No reasonable person could possibly interpret it as an actual threat. Most unreasonable people would even understand it was a joke.

Your example of gears is bicycles? That's cool, I guess. But, you know, *cars*...

I'm pretty sure most young people have also seen calendars, clipboards, film cameras, and Manila folders outside of a museum.

Unless things have drastically changed since (recently) I went to Caltech, math is its own department -- part of the division of physics, math, and astronomy -- and CS is its own department as well -- part of the division of engineering and applied sciences. There's also an applied/computational math under the E&AS umbrella, but that's not anything like either math or CS.

Generally coursework is the least important component of a graduate education -- in some cases, people won't take any courses at all during certain terms. Additionally, the courses generally require much more work, and a 'C' is a failing grade.

So you figured out what was going on and you didn't help your friend out?

Some friend you are...