One discussion that might be worth having regarding the crux of math (or other) education might start by imagining what real and highly leveragable things might be doable if some "philanthropic" funding were available to start it with. I suspect a lot of mileage is gotten by people who use publicly available resources such as mathworld and wikipedia and other math-oriented web sites, but the target audiences of those services are generally people who are already relatively well along in their math education. Given the existence and increasing pervasiveness of the web, its ability to reach such a large number of people, what kinds of services might it be possible to build to help kids along in their mathematical self-education? There's a lot that can't be done without personal tutelage, but there's also a lot that can be, at least for adult self-education. What about for kids? As John McCarthy said earlier in this thread, if it's boring, it isn't going to work. But presumably we also wouldn't have to solve AI to do some pretty interesting things either. Just wondering if anyone else here is thinking about these kinds of issues from the point of view of considering trying to bring resources to bear on them. Shel