I will soon be graduating with my PhD in mathematics, and my experience thus far as a graduate student in the US is that it's really hard to get a position in academia where your research is more important than your teaching, and it's almost impossible to get an industry job where you can research what interests you instead of what's immediately applicable to your company.

I won't have any trouble getting a job. I won't even have trouble getting a very well-paying job. But getting a job that lets me pursue my research interests as I see fit? Not without 2 - 5 more years of postdoc positions and some really great papers. And then 5 more years of writing papers like a maniac to get tenure. A lot of pressure to go after low-hanging fruit, even if it's stuff that doesn't interest you. That's a lot of years of 80 hour weeks just for the privilege of studying what you want.

I am a mathematician, and quite frankly I don't want you defending me.

Since the NSA just hands money to the AMS to be distributed as grants, I'm pretty sure it's all publicly available research.

Wait, are you addressing me specifically? If you are, that's decently insulting. If not, you sound like you're a bit off your rocker. Like a guy ranting on a streetcorner.

That's not the argument. Mathematicians aren't going to starve, but they're also not going to find many opportunities in industry to do mathematical research without direct, obvious application. Research that we have no way of measuring the value of, because applications haven't been found.

I agree that the mathematicians in the direct employ of NSA should take a long hard, look at their own ethical code, but the fact of the matter is, the NSA provides lots of funding for university mathematics departments. For research that is open to public scrutiny. From TFA, $4 million goes to a grant program administrated by the AMS and things like undergraduate research programs and number theory conferences. The NSA is just throwing money at mathematicians on the off chance that they discover something useful to national security.

If the AMS were to sever ties with the NSA, there goes $4 million of funding for public mathematical research in a puff of impotent outrage.

I'm all with you when it comes to not working for fascists, but we're talking about public research here, for the enrichment of all humanity. Not shady spying stuff.

Speaking as a PhD candidate in mathematics, while I personally won't have anything to do with the NSA (other than being on their watchlists, natch. I guess I shouldn't have dared to ever glance at Linux Journal,) I can't bring myself to hate on mathematicians who do. For all I know, my fellow grad students and I are only studying math because the NSA gave the university money to cover some of our stipends. In fact, that's probably the case.

Like it or not, if mathematicians cut ties with the NSA, there would be fewer mathematicians. Not just fewer mathematicians directly employed by the NSA, but also fewer mathematicians doing research at all and fewer mathematicians in training. American mathematical research would suffer a setback. I can see why the American Mathematical Society doesn't want that to happen.

I've seen a lot of love for i3wm, so riddle me this: what user-visible advantages does it have over other tiling window managers? I just keep finding stuff about how nice the code is.

I would take his claim as a hypothesis that requires further experimentation, not as bad science.

This is how science works. A scientist says, "I have a model that explains these phenomena in a way that agree with real-world data. It makes these predictions. Bring it." Then people collect data and do experiments to verify that the model and its predictions hold. Or, they discover discrepancies and refine that model.

The author has a model. He feels pretty confident about it. Now the science begins.

I teach out of Stewart's Calculus, and here's my "favorite" thing about the text. He has an extensive sidebar detailing the correct spelling of L'Hospital, and why we should honor the man by spelling his name the way he did instead of with the modern French spelling. And then he consistently refers to Johann Bernoulli as John.

You have citations for this? I find that really surprising, and I'd like to read up on it. Almost everything I read that isn't on my computer monitor is in portrait mode. Narrow columns with lots of linebreaks. Books, magazines, comics, my e-ink kindle, you name it. The one exception is highway billboards.

Also, what exactly is meant by longer and shorter lines? It seems to me like there would be a range of line lengths that make for efficient reading, and going outside of that range in either direction, too short or too long, would result in decreased efficiency.

I've seen the same sorts of complaints come up on /. again and again, but it's never those. It's always, "well if they don't want to, they don't want to."

Pair programming is a very social task. I'd hardly call that a tiny bit of edge dialogue. Citation needed on the bit about continued socialization all day in typical office jobs. Every office I've seen is full of people quietly doing their own work.