ecuzzillo

I should point out that the conclusion of the linked study is complete bullshit. It claimed that poor people are more polite than rich people. What it should have claimed was that poor people who went to Berkeley are more polite than rich people who went to Berkeley.

Poor people who go to Berkeley are likely to be moving up in the world relative to their families; their eyes are being opened to a whole new world of knowledge and power and all that, and they are humbled by it.

Rich people who go to Berkeley are likely to come from prep schools, and feel vindicated by having been admitted to a top school, and therefore are snooty.

The conclusion doesn't jibe at all with my personal experience, either, and I think this is why.

If you want to win at football, you most definitely need a good football coach, and the markets for pro football coaches and college football coaches are hardly separate. So you have to pay college football coaches somewhere in the ballpark of pro football coaches.

What has never been properly explained to me is the utility of winning football teams to universities. I really would like to hear about that.

In some ways. If you recorded just what the rods and cones detected, you'd actually get a really small, really high-resolution area in the center (the fovea) with the rest at extremely much lower resolution. So if you recorded what the quarterback saw, but without the post-processing done by the brain to cache some of the things recently glanced at, it might be pretty hard to watch.

What's the hobby?

It may be that refits require loans, and loans are hard to get now; moreover, the article may have been talking more about cafes in smaller, less busy towns, which banks may be less willing to lend to.

It could be that stages 1-3 mirror the company lifecycle. Companies start out not sucking (as startups) and slowly either become lousy and big, or get bought into a lousy and big company.

1. Make cool stuff. Examples of truly cool stuff include http://www.anybots.com and http://www.cs.cmu.edu/~adamwb/ (although stuff vastly less cool that that still counts as cool for purposes of finding a job)

2. Post it on your website in such a way that it is obvious that a) it is cool, and b) you are therefore the shit.

3. Cold-email (as opposed to cold-call) people you want to work for. If you have done 1 and 2 correctly, and the people you want to work for are in fact good people to work for, they will hire you.

It worked for me, and the stuff I made was hardly cool, and it was hardly well-posted on my website. I now have basically exactly the work situation I would define given complete freedom.

Yeah, we're doing it. We make humanoid robots.

http://www.anybots.com

How prodigious do you have to be for society to push you along like this? I accelerated a lot, because I wanted to get to the good stuff, and I felt burned out occasionally when I did too much pointless work unrelated to what I liked, but otherwise was fine, even in the heavier semesters.

Basically, I figure there are two reasons to accelerate. One, it's dumb to waste time on things that don't matter (like the two years of high school math over which are distributed about six weeks of material, also known as algebra II and precalculus); you wouldn't want to waste your life in high school any more than you'd want to waste your adult life.

Two, you have only so many years of real productivity, I'd say between about 18 and 39 (upper bound varies by field and individual), and you want to spend as many of them as you can doing real work that has an effect on the world, rather than education.

amazing link

It's also the next Unix in that it's kind of hard to use for newbies, and shits all over your data when you do something that seems like a routine action (git revert anyone?), but in fact will tell it to self-destruct. svn is emphatically not this way, which, when you aren't doing anything distributedly and don't have very many branches, is happy.

I've dealt with a couple systems that link Lisp 'trivially' to C. When they say they link trivially, what they mean is that they let you call the function in the C library just like you'd call it from C. This is frequently extremely not how you want to do it in Lisp, and so you're forced to either a) write a Lispification layer, in which you take all the C functions and turn them into a library that acts like you want a Lisp library to act (the usual approach), or b) just write it like it's in C, which makes you very angry all day long, and also loses some of the benefit of writing in Lisp. Both approaches slow you down quite a lot. Especially if you decide not to be an asshole and release your lispification layer as open-source (which you should)-- because then people will all be pounding down your door for fixes and extensions and maintenance for new versions of the C library, which is not what you want to do. You didn't even want to write the layer to begin with, presumably.

So no, 'trivial' C linking not actually a silver bullet making everything hunky-dorey with respect to lack of libraries. Python has a significant advantage in that it's popular enough for many people to have already gone through the Pythonification process for most important libraries, and have also open-sourced their Pythonification layers.

Googling "temperature berkeley, CA" works just as well, and it's built into your Firefox search bar.

Sturgeon's law strikes again.

Most accounts I've heard say that Microsoft was the place to be in the early 90's, and many smart people wanted to work there, particularly ones coming out of college, and Microsoft software was crap because executive decisions and backward compatibility made it crap, not because their hackers were crap. People who read the Windows 2000 source leak said that it was fairly uniformly well-written code, whose comments indicated that its authors knew they were making everything ugly and klugey, but couldn't do much about it and preserve compatibility.

So I somewhat doubt that good employee hackers can steer a fundamentally really evil main guy away from being evil.

Why do you put the names of languages that are clearly and obviously not acronyms in all-caps?

I don't buy it. Stress has nothing to do with getting kicked out of hack mode. Even the least stressful of interruptions, assuming it uses the local state registers that are used in hacking, can completely scramble your brain. The friendliest of comments from co-workers, assuming they're about hacking (but not your project) and require hacking thinking, can ruin it, if the comment turns into a conversation that lasts more than 60 seconds.

It's not about stress. It's about juggling a bunch of delicate state, and the state is being held in short-term memory registers that are easily clobbered.

Web 2.0 sucks in the sense that if you could be working on real things, it sucks to be working on two-week rails projects. Clearly, sometimes you have to do two-week rails projects or other equivalently boring things in order to get rich, but that doesn't make them any more interesting to talk about, from a hacker's perspective.

Web 2.0 sucks, I agree. Its suckage does not preclude bug tracker suckage, though. The GP had a point, which is that boring software is boring, and really, it would be better to read stuff about interesting software, by people who regularly write interesting software. There is interesting software floating around; the problem is, most of the people who write it don't have time to blog about it. They sometimes do write about it, but usually in that case it comes in the form of either patents or academic papers, both of which are unnecessarily hard to read.

The WMV format wouldn't ordinarily be a problem; lots of video players everywhere can play that fine. Even WMV streams are ordinarily OK. It's just that something freaky is going on with these particular streams.

Does it work on Linux? VLC and Mplayer don't seem to like the stream, yelling about connection problems.

Everybody who works in a particular field will say that'll be a big area of innovation. So, if you take people from a bunch of big companies, the answer you'll get is that the areas where there are currently big companies will improve. So if, in 1967, you asked everybody what the top areas of technology innovation would be in 1985, you'd COMPLETELY MISS personal computers.

Similarly, in my opinion, this is completely missing the advent of useful mobile, possibly domestic robots, which are significantly less than 18 years away from being viable, especially teleoperatedly (which doesn't require AI to be solved).

Right, but aren't they doing it for Linux or Mac?

When would you say the best hackers stopped building stuff on Microsoft platforms?

I recently met a very smart guy who has worked at Google for nearly a year, and he had almost nothing positive to say about it. In his whole large division, there seemed to be hardly anybody who actually got anything done. And, worse, there was no method for firing people who didn't do anything.

I don't think we're doing that much processing to do vision. It's never the case that even close to all the cells in the visual cortex are being used. It's always a tiny fraction of that. Yeah, we have a bunch of cells that neuroscientists think are somehow related to doing vision, but there's no evidence that they're doing some kind of horribly computationally intensive thing all day long. If they were, we'd see it on MRI's as soon as we put a picture in front of somebody.

No, I think we just haven't figured out what to do, not that we're limited by computational resources. I think there are a bunch of pretty strong assumptions hardwired into the brain, so we can see physical objects, but then you can fool us by showing us images that mess with the assumptions (aka optical illusions).

(Not that this is a particularly useful idea, mind you. I'm not about to solve computer vision. It's just a random philosophical idea.)

How about a link to that, then, on the top bar, since everybody wants search?

It doesn't work in the sense everybody most wants it to work, which is to slap a cheap CCD camera or three on top of a mobile robot and have it tell the robot everything about its surroundings, like human vision can. Yeah, if you point a camera at a very restricted specific set of objects, all lit the same way, and all in generally the same basic configuration, you can do some vaguely useful statistics to get some information out of it, but that's kind of a painfully limiting thing.

Put it this way: At Anybots, we have a robot with 16 cameras on its head. We pump all those cameras back to an operator station, stitch them all together in a vaguely fitting-together way, and then project them onto a big bank of monitors. Plop a human operator in front of the monitors, and he knows exactly the orientation of the robot, the configuration of its limbs, and has basically a 3d map of the entire scene and a model of the lighting, minus occlusions. I'm pretty sure nobody knows how to write a computer vision system to do that. I'm not even sure that anybody knows how to use the overlap in the cameras to figure out how to make a coherent nice projection of the cameras onto the monitors.

The reason there isn't a definitive text on vision is that vision doesn't work. Algorithms work, and so to have a definitive text, you just collect up all the notable ones that work, and write about them in a book. When you have a totally unsolved problem like vision (outside of certain highly constrained domains, like robot soccer, where the colors are guaranteed, or industrial tasks, where basically the whole scene is guaranteed), you basically have to choose which methods that currently don't work might have a chance of eventually leading to something that does work. There are many possible sets of choices for that, which is why you can't really have a definitive text.

Sturgeon's law: 90% of everything is crap.

This applies especially to attempts at NLP.