well first you gotta take some marketing classes, then some psyche classes.
people will always pay for escapist software. the cooler, the better. thats why doom3 will be such a hit, because its so un real that its cool. of course unreal really did beat out id software, and always will be the number one game, but thats because it was truly so immaginary.
well once you figure out what people want, then you can follow the companies that deal with applications of their software. as much as our lead coders hate to say it, I think that ID sofware has one thing that the others dont, a real-world applicion of physics and that's why the game does so well. Yea air control and some vector stuff like non-additive velocity on rockets tend to play with your head, but tribes and v2 don't even come close.
Watch the boards. There is a big, huge rift right now about lightmaps within bsp's. BSP means binary space partition, was developed at MIT in the 60's. Basically it was 'out of sight, out of mind' for computers. Doom3 isn't going to have the same concept, they are trying to pass some of that bsp work over to graphics cards. IMO this isn't the way to go, I mean its never going to be perfect no matter how phat your system is, might as well set a 'mood' and run with it.
Hence WFA 3.5, we are really the forerunners in this new medium. Some of our levels will stress people's cards in lightmap, that we already know. But in vertex, it's going to be faster then the 3.3 maps were; thats a fact. (skyportals not considered).
Console games are the coder's wet dream: develop for one system/one platform only. No worry about unsupported hardware; what works for one works for many and no reason for 'patches' or 'updates' to keep up with that. You produce a game, make a few bucks, move on. I would love WFA to be a console game, it would be so much faster then what you see now but god damn what acrid and static and bp did with the code defies immagination.
Just learn what makes a game a game. Visit all the openGL sites, subscribe to all the mailing lists. There is always alot of talk, and alot of people who will help out. To me, this is a hobby. It's getting to the point of interactive lightwave and 3ds, to me thats something just fun to do. To others, its a matter of working your way into the fray and selling yourself, "pushing pixels" to the mainstream.
Look at nintendo. You could get a job, if you know how to do documented testing in chemical engineering but never played a game, just by beta testing their games. All the industry wants is cogs, and it depends on where you feel comfortable setting yourself in.
All of us people who do it for free, well, they aren't chasing a dream, they are actually just plain old better then those who do it for money because they themselves are doing it for their own reasons: to make a kick ass game for people to enjoy. Thats what makes WFA special, and why we never will go retail. We listen to people, we work with people, we really are the most unique mod ever because the developers actively envelop themselves within their creations, they don't just dish it out for acceptance or pride.