[callahad]

There's an interesting paper from a few years back looking at software defect incidence in Windows Vista and trying to figure out what models did best at predicting error-prone modules.

Turns out, in that case, organizational structure had better precision and recall than any of five other models (code churn, cycolmatic complexity, test coverage, etc.)

"The Influence of Organizational Structure on Software Quality: An Empirical Case Study" -- http://research.microsoft.com/pubs/70535/tr-2008-11.pdf

[jacksonh]

This is awesome. Thanks. Love this kind of stuff. If you are interested there is a book "Making Software: What Really Works, and Why We Believe It" that takes a scientific approach to evaluating software practices.