I think it's time to lay down a couple of things once again very clearly. I don't want to boss around, but the fact is I do have a lot of experience with projects of all kinds, many of them led by myself over the last 10 years. Some were finished, some failed. And I have to say the way things have been proceeding in the Korax project these last 6-9 months (or better said, the way they were getting out of hand), is the surest way to losing it all and making sure the project will never see the light of day.
A project normally consists of several phases, in which planning clearly comes before doing. This also means in most cases that once work has commenced in a major way, new ideas (unless small things) shouldn't be considered at all any more.
There is also the bigger picture to consider. There are currently 3 different projects on our roadmap: Scattered Evil, Korax Arena and The Serpent Power. You all seem to be focusing so much on SE as the project and keep forgetting SE is more of an evolutional dead-end and a kind of test of the whole principle (after all, in some ways it'll be the first game of its kind) than the final step. That will be TSP.
That means in the case of SE it's more important to bring it out ASAP than to polish it until utter perfection. If you are still not convinced, think of Daikatana... 4 years in work by an industry mastermind and in the end not only not worth the wait, but even compared to games finished within a year a major letdown.
Another thing about SE is the engine: it's utter crap. I want to leave it behind ASAP and that means releasing SE. One of my problems with mapping was that I felt my hands be bound by Doomsday all the time - something quite different in the case of Vavoom.
Something else... fan interest in Korax has been dwindling over 2002 - very clear why: no new releases of any kind for nearly a year. This would all change with Korax Arena - which is delayed because of SE.
Don't misunderstand me, I don't want to push out SE tomorrow, no matter how bad the quality... but I want it out in the first half of 2003, with a minimal set of features over KMOD2.2 (which I'll list shortly, now I've got to go).