turn-based or non-fps gametype

#1 Seriously - I began with pencil and paper.

I thought about what I wanted to create and began sketching out some ideas for how the world would look, how the buildings would look, and what the user interface would look like. Nothing fancy.

Then I started looking at the demo to see what it had already and noted what I wanted to change - background screens, character creation, and the like, then I went to the scripts and figured out as much as I could about how they worked and how I could change them.

Next I went to the forum and asked a bunch of newbie questions about how specific things worked. I read the replies, the appropriate forum tutorials, and started making changes.

Finally, when thought I had most of it figured out I revisited my original idea and measured it against what I knew the engine could do, what I knew my team could do, and what we might be able to do with a bit of luck. Then I built an initial game design doc and showed it to my team and started haggling over what we could accomplish. They tore it apart, improved it immensely and now I'm finalizing it into something we can all look forward to playing.

I have a small team that has a lot of raw talent and gameplay experience but is weak in coding - so I try to stay out of the regular code as much as possible and do as much as I can using scripts. We chose to emphasize original and improved gameplay over new features.

For now, I would suggest you try the scripts first and go into the code later - for the reasons above and because you will be way ahead when the CVS comes online. You will not have to undo or merge your code with the improvements in the CVS code.

Hope that helps.