Jason dropped a post on SVN today about the importance of the “middle” of the development cycle that worth a read – actually, I think that it should be a MANDATORY read for anyone in a position of power in the development world but I digress…
I’ve fallen victim to this myself more times than I can count – you’re all pumped about a new idea/project and get right to it. You plan and diagram so as not to miss anything. You start to throw down the code, pumping out beautiful, clean code complete with unit tests and XML comments. Initial revisions are flying thru the automated build system. Then somewhere in the middle, things start to cool down. The comments come in a trickle, if at all; unit tests are a pain so each class gets a little less built. Often times there are small surges when a class will get built here or there, maybe a little more documentation gets done, but usually there’s a big stall out in the middle that’s followed by the disappointing realization that the knowledge gained thru the coding process means that the original design needs to be reworked/tweaked. So where do you go?? Back to square one thus negating efforts to-date?? Or simply push on to the initial version and figure that you’ll return for a redesign/refactor for the “next version”??
