So I've been sitting here, thinking, about some of the projects I've been working on. I'm really enjoying working on my PyBoxel project, things are actually starting to come together on it. I've got a firm grasp on how I want the project to proceed, and I'm actually spending time working on it almost every day. I think it will be a pretty big hit when I roll it out and really start getting feedback on it.
I've also been thinking about some of my other projects, and I think if I can get them out of my head on paper here, that I can leave them alone for now to keep working on PyBoxel.
Building a game console in Linux has always been a long term goal of mine. There isn't a whole lot to really do with it, other than make a gamepad friendly window manager. Everything else is already practically written. Aptitude with package signing provides secure content distribution method, wireless drivers, display drivers, etc, it's all there. The Devkit would basically be SDL, and a hand full of other small libraries.
Now that I have the spare PC guts, it might be worthwhile to start building a prototype. The D525 isn't much, but it's more than enough for most games.
Another thing that I've been thinking about is that processing game input sucks. I think a modular input server would be a fantastic idea using raw input, allowing the client to handle interpreting the byte strings would be just the thing. It's a very scary sandbox to be playing in, but I think with the cross platform libusb, it might be just the ticket for raw-reading pesky usb game controllers that might not otherwise have drivers.
I've also been thinking greatly about how to make programming in python more accessible. Thinking back to the elder computers such as the C64 and the TRS, instead of spending hours playing games in-front of a TV screen, you could handily make your own in some dialect of basic. I think a portable everything-but-the-screen setup would be a lot of fun to work with.
I can start putting together the right Debian spin or whatever, but finding the cheap hardware, Ideally something around $100, with like an atom or a geode CPU would be idea. Bundling it with a book would be a good idea too.
http://www.commodoreusa.net/CUSA_VicSlim.aspx
something like this, but with a cheap ARM cpu. I think the raspberry PI would make a great foundational component for something like this. until that comes out, the beagle bone is only 89.
So, back on track with the pyboxel project then!
