Identifying Features in Forks

Fork-based development has been widely used both in open source
communities and in industry, because it gives developers flexibility
to modify their own fork without affecting others. Unfortunately,
this mechanism has downsides: When the number of forks becomes
large, it is difficult for developers to get or maintain an overview of
activities in the forks. Current tools provide little help. We introduce
Infox, an approach to automatically identify non-merged features
in forks and to generate an overview of active forks in a project. The
approach clusters cohesive code fragments using code and network-
analysis techniques and uses information-retrieval techniques to
label clusters with keywords. The clustering is effective, with 90 %
accuracy on a set of known features. In addition, a human-subject
evaluation shows that Infox can provide actionable insight for
developers of forks.