Coordination and productivity issues in free software: The role of brooks' law

Abstract

Proponents of the Free Software paradigm have argued that some of the most established software engineering principles do not fully apply when considered in an open, distributed approach found in Free Software development. The objective of this research is to empirically examine the Brooks' Law in a Free Software context. The principle is separated out into its two primary premises: the first is based on a developer's ability to become productive when joining a new team; the second premise relates to the quality of coordination as the team grows. Three large projects are studied for this purpose: KDE, Plone and Evince. Based on empirical evidence, the paper provides two main contributions: based on the first premise of Brooks' Law, it claims that coordination costs increase only in a very specific phase for Free Software projects. After that, these costs become quasi-constant. Secondly, it shows that a ramp up period exists in Free Software projects, and this period marks the divide between projects that are successful at engaging new contributors from others that only benefit from occasional new contributors.