One thing I wanted to do as part of Spaghetti Code this year was to have some sort of project that I could dink around on over time, possibly host on CodePlex, and do little screencast shows as I built the thing.  Problem is, I have a tough time coming up with a project that is a) interesting to people and b) interesting enough to me that I would try and keep with it.  I had tried the Overlay project a couple years ago and it went nowhere.  Mostly because it was not all that creative - it was basically combining a bunch of tried-and-true techniques to come up with a larger concept.  Whatever - I got bored.

So what to do now?  A game sounds fun.  If you have ever played Star Fleet Battles (the board game, not the computer versions that are out there) I kinda liked that game.  Not so much because of its Star Trek theme, but more because I always felt like I was playing a submarine battle in space.  I loved all those submarine computer games back in the day - Silent Hunter, 688, or whatever they were all called.  My biggest problem with SFB was that it never could mimic the the "blindness" of a good submarine game.  I love the stalk and prey aspects of a good submarine game.  I'm thinking of something like that for a hobby project.  No heavy graphics - just a great UI that lets you prowl the deepness of space using sensors, probes, and your wits to hunt down and destroy an opponent.  Mutli-player support would be the only way to go.  And while it wouldn't have a lot of heavy graphics, it would certainly need some good models for sensors, environments, homing devices, etc.

It sound fun to me, perhaps even fun enough to keep my fleeting interest maintained for a while.  WPF or DirectX is the first decision. Hmmmm....  I know WPF better than DirectX but some of the sensor display ideas I have seem to lend themselves to DirectX. 

Thoughts?