Description
When we build multi-networks, have the possibility to combine various types of networks (author networks, artifact networks, bipartite networks). However, commit networks are not possible to be part of multi-networks (yet, for good reasons).
To be able to incorporate the data of commit networks more into a more general network representation, constructing bipartite networks that combine authors with the commits they have authored (or committed) might be a first step.
Motivation
Combining authors with their commits could be useful for multiple reasons:
- One could use weights of connections between commits to predict the strength of connection between authors.
- One could map multiple authors to a commit (at the moment, only a single author and a single committer, but maybe more in the future, given that also additional people could contribute to a commit via suggestions, etc.), which provides the possibility for a new branch of analyses.
Ideas for the Implementation
Construct bipartite networks between authors and commits in a similar way as the bipartite networks between authors and artifacts are constructed.
Additional Information
We should start with authorship relations first. When considering committer relations also, this might be related to #84, which would need careful consideration.
Description
When we build multi-networks, have the possibility to combine various types of networks (author networks, artifact networks, bipartite networks). However, commit networks are not possible to be part of multi-networks (yet, for good reasons).
To be able to incorporate the data of commit networks more into a more general network representation, constructing bipartite networks that combine authors with the commits they have authored (or committed) might be a first step.
Motivation
Combining authors with their commits could be useful for multiple reasons:
Ideas for the Implementation
Construct bipartite networks between authors and commits in a similar way as the bipartite networks between authors and artifacts are constructed.
Additional Information
We should start with authorship relations first. When considering committer relations also, this might be related to #84, which would need careful consideration.