R. Stuart Geiger.

Bottom Line Up Front

I rate this 7.7. Does what it says on the tin.

Summary

• 5500 randomly sampled respondents from 3800 github repos.
• 500 non-random, non-github responses.
• 50 questions.
• paper is really an ipynb https://github.com/staeiou/github-survey-analysis/blob/master/github-survey-descriptive-stats.ipynb
• They cite Python 3.6 as a 1995 paper by guido? Also a bizarre subset of all the ipynb greatest hits all as papers. That’s a first for me. Truly bizarre.

Amos’s Thoughts

It’s mostly a bunch of figures. I will now just vomit my uninformed opinions at first glance of any figures I feel like commenting on.

• 4.1.2 slightly more active members than the pareto distribution I’d expect.
• 4.2.1 very positive skew. Blows my mind License is more important than active.
• 4.2.4 nobody finds open source harder than proprietary?
• 4.2.7 self identification ‘i am part of open source community’ could be an interesting split
• 4.4.1 received help, mostly yes, interesting
• 4.5.3 ‘I am not permitted to contribute at all’, I wonder if these ones do? Small sample size though (63).
• 4.6.4 Literally zero women. Just like wikipedia, this never fails to surprise me. My maleness is showing, but… why? What are women doing? There’s no obvious wall? A quarter are contributing totally anon…
• 4.7.2 Lol only half have ever seen rudeness?
• 4.7.3 No experience with rudeness

Okay, well, the end seems to be a lot of small numbers about blocking users/going to the law. Overall pretty boring set of data. I’d loved to have tried to learn more about these users, what projects, not just aggregate ‘do you use your name’? Anyway, now I know.