Make a demo.
Then worry about whether you have to figure out the legal stuff.
Till you make a demo, it's all vapourware.
I'd suggest demoing it in an established engine, quickest route to getting something playable. While I don;t agree with Descent 4's choice of Crystal Space, if you are HALF the programmer you think you are, you should be able to get something playable out of it in relatively short order. That gives you something to show, proof some of your playability thoughts, while you code your own engine.
But coding an engine THEN making a game out of it is
of work. I went the "license" approach and shelled out the 100 bucks for Torque, so I didn't have to start from the ground up while learning the game related stuff. Better documentaion, more full featured engine (out of the "box"), already cross platform. Another good engine, that can have playable stuff in weeks, buys you time to code your own or come up with the design documentation, or whatever.
Demo, then hard core code. The D4 people did some concept stuff using the D3 engine. That was one thing I think they did right, got something playable in peoples hands quickly.
Watch the "god" complex, and good luck.
