danbri / cg2xyz

W3C CG data extraction tests

cg2xyz

W3C Community Group data extraction tests

• extractGroups.py - reads W3C Community Group site, writes local JSON summary
• toGephi.py - reads local JSON, writes CSV file for gephi.github.io analysis

The idea is to look for ways of organizing/analyzing the many Community Groups at W3C. See also http://www.w3.org/2015/10/factsheet.html#Groups for high level overview.

At this point it is a proof of concept that shows crude extraction of a group ID to Person-name bipartite graph. The gephi code is also crude, it just counts the number of people's names that are in common between any pair of community groups.

Not handled:

• extracting user IDs instead of names
• extracting photos/icons, group descriptions, etc.
• extracting links to mailing list(s)
• any freshness/aliveness metrics
• extraction from associated public mail archives, e.g. active contributors