There is a good description of factorial development, which grahamp mentions above, in Ansel Adams' _The Print_. The idea is to be able to compensate on the fly for changes in temperature and exhaustion of the developer over the course of a printing session by maintaining a constant ratio between the time it takes for the image to emerge to the time it takes to complete development.
For instance, when I develop Azo in amidol, the emergence time is about 1/3 the total development time. Adams described using a metronome for factorial development, which I've done, but I usually just leave a timer on and keep track, when I'm working this way. It's also useful to know the emergence time for water bath control, since this is when I'd usually pull the print from the developer and transfer it to the water bath.