Wizard Schools?
Steve Yegge wrote a nice exposition about private supergeek boarding schools. If you haven't read it, stop now and read it. If you continue on without reading the original article, you've been warned. :)

...
...
...

OK, so Steve's post was a mind experiment, but rather interesting one. First of all, I (and other readers) caught ourselves thinking: "How did we miss this thing? Is it going to make us obsolete?". But then, when we realized that there were no such schools in existence yet, the more interesting question became: could there be? It's impossible to answer this question without trying to build such a school. And anyway, those who can do, those who cannot say "it's impossible". :) But here are some thoughts about wizard schools:

Steve assumes that the quality of kids from Wizard School will be dramatically better than the ones coming from ordinary schools. Unfortunately, this does not seem to be confirmed by historical data in computer science. Yes, MIT and Stanford and CMU (and few other top tier universities) graduate students that are on average better than the ones from second tier universities. But really they are mostly just somewhat better. It's not as if Stanford graduate will just run loops around U of Kansas graduate. Statistically Stanford graduate might be more hard working and talented, but that's pretty much it. So it's not really a proven fact that Wizard schools will do much better.

"But wait" - you will say - "Wizard schools will go through very demanding selection process and they will start much earlier than universities". OK, let's address these two things. It's not clear what age Steve intended the kids to start at Wizard Schools. If he intended them to start at ages 7-10, then the thorough and demanding selection process is unrealistic. Kids at that age don't know if they want to be geeks and sending them to a geek school might be a power trip to parents, but it's not guaranteed to be something the kids want. Also, although there are child prodigies, some of children show great geek material much later than 7-10. So by thorough selection we might throw out babies with the bath water (pun intended). If schools will start at later age of 14-17 or so, then the selection process is more realistic, but then the school becomes more similar to a college. And it might be too late to teach kids wizard techniques to program closures in their sleep.

Finding teachers and great teaching methods is another hurdle. It's not as simple as selecting the inventor of Perl, Python or something else. Teachers would have to spend almost all their lives teaching the kids. How many university professors or industry practitioners would select such a job? Even if it came with comparable salary to their current one? How many of the ones who would want to do it, would be good at it?

There are currently a lot of private boarding schools around the world. Some of them list quite a few famous alumni, mostly in politics and arts. It seems that the main attraction of these schools is small classes, high student/teacher ratio and possibly connections network for the rest of your life. I assume some of them also try to bring out best in their students, to make them into Wizards. Since we have not heard any great success stories, it seems that few succeed in that quest. It seems that they are mostly Well Educated and Well Connected schools, rather than Wizard Schools.

Does this all mean that someone should not try to create a Wizard School? Not at all. It's a fascinating idea that's worth trying just because it is fascinating. We are talking Wizard stuff after all, aren't we?