This is an interesting comparison, but it is most likely just a high overlap between their userbases - people that use github also use stackoverflow, and vice v

This is an interesting comparison, but it is most likely just a high overlap between their userbases - people that use github also use stackoverflow, and vice versa. This also doesn't include the number of people programming with a language in a corporate environment whose work is not public nor on external servers. Megan Squire posted in a comment this link to a collection of multiple source code repository websites datasets, although it is only open source projects: http://code.google.com/p/flossmole/downloads/list http://code.google.com/p/flossmole/downloads/list