The Yard and Heron projects are very similar to a class I need for my Computer Science degree (compiler design). But the Yard and Heron projects ask the kinds of real-world questions that are glossed over in a classroom. For instance, Yard's currently debating whether to use static data like Chris Diggins originally designed, or references like the other programmer on the project has suggested. (To be honest, this is work that I won't be actually doing, the other programmer will likely be the one working on it; not to fear, though, I'll stay busy).

A project that I want to do, that happens to be similar to another class I need to take (operating system design; yes, it's the class that led Linus Torvalds to write Linux), and that I haven't started yet (and that will be more useful than the class because it will involve real-world questions and answers) is the eCos Template Metaprogramming project. This isn't really the right kind of blog to talk about the details (in fact, I'm likely to start a blog for that when I do actually get around to starting the project). Basically the idea is to take a ten-year old operating sytem, and modernize its internals to make it easier to update. eCos is currently distributed by Red Hat.

However, while technology's getting more complex (making it harder for one person to pull off the next big thing), strategies for dealing with that complexity are becoming common knowledge (making it easier for one person to pull off the next big thing). Overall, we get cooler products each year.