mark-t

I think ideas like "Oh, that guy was a genius. Normal people could never do that." are fundamentally misguided. He worked tirelessly on mathematics, filling notebooks in his spare time for the pure joy of it. If you can find somebody else with that kind of devotion to mathematics who is hopelessly average at it, that would be something.

I mentioned Ramanujan because he's famous, and he obviously didn't get it from taking university classes. I used to do math competitions in high school, and I'd meet the best students in the state or country. You know what? All of us learned the same way Ramanujan did: we read number theory books for fun, continually asked questions that were interesting to us, and then worked on them until we answered them. Not a single person in the group learned by attending lectures or even through supervision by experts. It's just not how it happens. That's the point I was trying to make concisely before.

Mathematics can certainly be learned outside of a university. Look at Ramanujan. Or go find any math professor at a decent university and ask if the majority of his/her learning happened in the classroom. You'll get a resounding no, if not laughter.

Sometimes. Many parts of math are very difficult to tell whether you're wrong or not, except by reading through the proof extremely carefully. For example, about a hundred years ago there was a proposed proof of Fermat's Last Theorem, but it had a very subtle flaw: it implicitly assumed that unique factorization held in a particular set of rings. This turned out to be wrong in some cases. This error was very common back then, and many people believe this was the proof Fermat had in mind.

So, you see, in mathematics your unconscious assumptions are sometimes wrong, and there's nothing that would tip this off to you, like a compiler. In fact, very few areas of mathematics are the sort where you can easily tell when you're wrong.

I suspect you think of your high school math classes when you think of mathematics. That's really a very poor representative. Mathematics is about proofs, not answers. To be fair, though, the original post makes a similar misconception. Computer programming : computer science :: high school math : mathematics.

> Most solar cells in use today are either photovoltaic, meaning they harness only visible light, or thermal.

Hmmm. I'm pretty sure that's not the definition that I learned.

I went to Northwestern, and I must say that "(a Big Ten university outside Chicago)" is not how I would describe it. It's the only private school in the Big Ten, and it's dwarfed in size by the others as well. To call it a Big Ten university is about as far off the mark as you can get while still being accurate.

My sleep schedule drifts forward gradually. Right now, I go to bed around 6pm and wake up around 2am. Tomorrow, it'll probably be an hour later. A couple weeks from now, I'll be going to bed at 6am.

I do find that I'm much more productive when I'm awake during the night. It's probably because there aren't any interruptions and I couldn't run an errand if I wanted to.

I think the bar for a non-programming, non-startup article to count as "interesting to hackers" is a lot higher. Really interesting science articles just aren't written as often as somebody blogs something useless about how great closures are.

Your English is fine, by the way. I wouldn't have known you weren't a native speaker if you hadn't said so.

The logic is sound if you take the stance that phishing is the biggest threat to security. I don't know whether it is or not, but XSS and CSRF at least are under the website developer's control. Malware and phishing are the head-scratchers for people who know how to build secure websites, and the information card system does partially address the phishing problem. It prevents the attacker from being able to log into your accounts, but it doesn't prevent you from telling him lots of other things.

The trouble with math is that the exceptional mathematicians (including theoretical physics/computer science) are the stereotype, and the regular mathematicians are actually quite different from that. So in fact it's the stereotype that's wrong, not the claim that exceptional mathematicians are not very mathematician-like.

"Ask any independent software vendor what he hates most about developing for Linux and he'll tell you that it's having to develop for SUSE and for Red Hat and for Ubuntu and ... you get the idea."

Nope. He lost me right off the bat. Maybe I just don't use strange libraries, but getting things working across the Linuxes has never been a major problem for me, apart from debian/ubuntu's broken menu system, and this app sure won't fix that. And it's certainly not the available libraries that cause problems. The things that bite me are the wider unixes: Sparcs have numlock as mod3 instead of mod2, BSDs don't use GNU make by default, Apples have one mouse button, cygwin can't store much on the stack, etc. And honestly, compatibility issues are maybe 1% of the bugs I fix. This is no silver bullet.

But if I had to say something about the disparity of Linuxes, it would be all the different compiler versions and flags that people use.

Either it's running too fast on my computer, or this game is nearly impossible. I have a feeling it's a little bit of both. For example, I have to save the game every time I'm about to approach an enemy, as it will invariably take half my life about 90% of the time, so I have to replay it over and over in order to have any chance of getting through a room.

There's no question as to the feasibility. The desirability is another matter. I would sooner write TCP packets by hand than use that thing.

Actually, I do have trouble with Gimp, too. I have to google how to make a straight line every time. Choosing brushes and changing the color are a pain, too. But I don't use it often, and about half the time I'm using it where I would have used MS Paint when I used Windows. There's probably a toy program somewhere that fills that gap, and I just haven't bothered looking. Maybe I should pick up Xfig.

You know, I just don't see it. Maybe it's because I'm a developer, but I just don't see that many usability problems in open source projects, and these articles are going out of their way to avoid giving concrete examples. I'm marking this (the MPT article) down as Microsoft propaganda until the "interface design experts" start pointing things out.

It's not like it would be hard. Configuring Xorg makes me cry, and CUPS was still no walk in the park last time I checked.

Are you suggesting religions don't need to be defended?

I thought wasting time was the point.

Seeing as how the Python and Tcl interpreters are both written in C, this really just seems to imply that the C code wasn't written very well.

You need sed -nf.

And if you want to make it actually playable:

while (sleep .5); do xvkbd -text "\r" 2>/dev/null; done & sed -nf sedtris.sed; pkill sleep

(Note: I tried just using echo with a pipe but couldn't get it to work. That would naturally be a better solution. If you can get that working, let me know.)

Ask around. 16 is quite late to start taking math seriously and end up at the very top. He wasn't doing math for fun. He wasn't asking himself mathematical questions. He wasn't doing math competitions. He had no knowledge outside of what was taught in school to students at his grade level. This is freakishly uncommon, but he shows it's possible.

25 is young. He is just a few years into his PhD, and he hasn't made many major achievements (he's been a grader for the IMO, written one paper that's still being refereed for publication). The point is that he's noticeably better than I am at almost every branch of mathematics except combinatorics, and he had no interest until he was 16.

Math education is so bad right now that any genetic differences are drowned out almost entirely. Of all the mathematicians I've met (including IMO gold medalists, Fields medalists, etc.), there seems to be little correlation with siblings, at least beyond what you would expect from having parents who emphasize learning math.

To answer your question, if you want to contribute significantly, you will need to study hard for several years, but it is possible. One of the best mathematicians I know well didn't have any interest until he was 16, and he's now 25 or so.

I've watched a lot of the ArsDigita lectures on archive.org, too: http://www.archive.org/details/arsdigita

Awww, I was really expecting something interesting to happen after I clicked on the "Please do not press this button again." button.

Oh, and by the way: http://en.wikipedia.org/wiki/Klaatu_barada_nikto

When I find that I visit a site that often, I subscribe to the rss feed. I use keywords for searches: / for google, m-w for merriam-webster, and so on.

To be fair, there is an obvious benefit to being able to search titles and bookmarks. I've been using the betas (and nightly builds) for quite some time, and it's stopped getting in my way. I do believe that the default behaviors were not thought through very well (strings at the beginning of a url should have higher precedence than page titles), it does take up too much space, and it takes too long to show up. It also tends to interfere with autofill.

Actually, it would result in one person giving a reason and a bunch of people upvoting that. Whether or not that's a good thing is still up for debate. I do think people should feel compelled to give a reason, though.

Journals are really picky. The whole process takes ages. I'd recommend just writing it up and posting it to the arXiv.

I wholly agree with the principle here, but I often find myself running up against firefox. For instance, there's this: https://bugzilla.mozilla.org/show_bug.cgi?id=341067 . You'd think if that many people find a behavior unintuitive or undesirable, the developers might want to consider that there could be something to it. Of course, mom won't ever run into this because she'd have to change a preference to encounter it. It is possible to be too focused on casual users.

Just because two planets have been around for the same amount of time in similar conditions doesn't mean they should both exhibit life. The emergence of self-replicating molecules at any given time may be extraordinarily unlikely, such that the expected amount of time for it to happen is on the order of trillions of years. However, when you've got several planets in the same circumstances, one of them will likely succeed much sooner than the others.

Here's a more concrete example. Consider the odds of rolling a 528 on a 1000 sided die, and you have 5 people rolling such a die. The expected number of rolls for any individual to get a 528 is 1/1000 + 2 * 999/1000^2 + 3 * 999^2/1000^3 + ... = 1000. However, the probability that at least one person has rolled a 528 passes 50% after n rolls, where (999/1000)^(5n) = .5, or n = log(.5)/(5log(.999)) = 139.

In other words, you expect to have rolled a 528 after 1000 rolls on average, but you expect somebody to have rolled one after 139.

That's what a noscript tag is for.