Matt Assay points to Alan Cox’s talk - Dear Mr. Brooks, or: Software engineering in the free software world. There are not many talks that are informative and equally entertaining to engage you through their entire length. And this talk is compelling enough to even make the transcript quite enjoyable.
The talk has some enlightening facts about open source projects, and how it contrasts the traditional proprietary models in many aspects. Using Fred Brooks’ The Mythical Man Month as a base, Alan shows how the open source projects deviate from the ways of the traditional software development. It is not that the software engineering rules have changed, it is more about which of them take precedence and dynamics of the team.
One observation I have made over the years is that our software projects are too dependent on the estimates, which are not very accurate. Estimates are estimates, they do not tell us when it will happen, but when it might happen. Unfortunately we ignore the difference for the convenience of our project management operations and bear the cost. Not that estimates are not important, but it is a big mistake to start considering estimates as schedules.
A very informative talk. Even if you do not believe in open source, go through it, you might find gems which might help you improve your development model.


October 25th, 2007 at 6:06 pm
The second link(Dear Mr. Brooks, or: Software engineering in the free software world) is broken.
October 25th, 2007 at 6:32 pm
Thanks for the catch Binny, it is fixed now.