But adding coders to a long term project will, in the long run, accelerate development enormously!
In the long run, everyone's dead.
Most large multi-person software projects are set up, from the very beginning to have a workflow involving, y'know, multiple people. They also are developed using methodologies that have been designed to make easily maintainable code. None of these things apply to DF, which also happens to be huge: hundreds of thousands of lines at least. In order to have other coders participate, it would first be necessary to re-organize the code, a monumental undertaking which Toady would have to direct and during which no development could take place.
Then you have to get the other coders up to speed on the project. But wait these are "volunteer" coders who don't work full-time, right, so getting "up to speed" on a project which -- did I mention this already -- is already hundreds of kloc, is going to take forever. Getting Toady to train them to not do everything wrong is going to take another forever. And during all this time Toady is going to be able to do basically no development because he would be doing all this other organizational crap (which he has repeatedly stated he is uninterested in) which would take up all his time.
You would start getting real productivity gains in one to two years, if Toady went all-out to integrate the new coders into the project. To make up for the lost time and make the whole enterprise worthwhile would take several more years. Who wants to wait that long?