Microsoft's development process

I just became aware of the Engineering Windows 7 Blog, and I have to say reading this post brings a lot of things to light for me. Larry Osterman, the author of this post and a long-time Windows developer, compares his experiences with Windows Vista and Windows 7 development. There are a few points that I found interesting: firstly, "All-in-all, this [Vista feature development] wasn’t too different from the way that features have been developed at Microsoft for decades (and is basically consistent with what I was taught back in my software engineering class back in college)." In such a rapidly changing industry, I would think that it would be best for companies to constantly adapt their development techniques to match the demands of the industry. No wonder so many Microsoft products turned out like they did. Pretty much all of the remarks I have stem from this idea.

Microsoft continues to use the waterfall method of design, which was hammered into me as a Bad Thing for the entire semester of my software engineering class (very recently). I may be incorrect in assuming that most other universities teach the same principles of software engineering, but I am going to make it anyway. I would imagine that the majority of companies now use some form of iterative development process. It seems straightforward that if it is taught at universities, used by a some of the most innovative companies and advocated frequently in Dr. Dobb's Journal, it is a big improvement over the waterfall method. If Microsoft wants to put out good-quality, innovative products, they should jump on the iterative design bandwagon, too. Clearly, not doing so yet has not worked well (judging most recently by the quality of Windows Vista and the amount of time it took to develop).