I was led to this article by an internal mailing list at Intel that centered around Agile Development. Agile programming, particularly XP, is gaining popularity within Intel and many teams have started to find greater success using it rather than the other methodologies that had been practiced. I am sure everyone has their own explanation for why that might be so – and many of them may try to explain it by saying that the process is light enough to allow them to do their job. Others will focus on the increased emphasis on communication as the driver for success. Cockburn points to communication as a major driver, but for a different reason.
Cockburn identifies the inherent inconsistency that we have as humans as being a major failure point with projects. He then identifies frequent communication as the factor in driving out that inconsistency.
Being good at communicating and looking around counter inconsistency, leading to the prediction that methodologies can make good use of low-precision artifacts whose gaps are covered by personal communication. Project histories also support this prediction, subject to the normalization of an adequately skilled staff, including management.
So the frequent and close proximity communication that XP promotes drives out the inconsistency in the project participants allowing for great success. In the article he talks about his struggle in trying to explain the success of projects and that he finally concluded that process was subordinate to people in defining the reason for success of a given project.