Developer Satisfaction Survey/2021/Categories/Code review
📖 Developer Satisfaction 2021
The developer satisfaction survey is an annual survey of the Wikimedia developer community.
Code Review
editSocial vs Tooling
edit💬 "Satisfaction with Code Review" seems too broad of a statement. Not sure if you are asking about social workflows or specific software tools.
Comments mention social and tooling issues. It's unclear whether the 52% overall satisfaction is about code review tooling, social dynamics, both, or neither.
Room for improvement next year.
38% of comments mention social problems in code review
50% of comments mention tooling problems in code review
Developer Equity
editwhy are people not reviewing my patches :(
– Volunteer Developer
21% of comments were about "volunteer equity". Staff's perception is that it's harder for volunteers to get review. Volunteers said it's hard to get review. Both comments are "volunteer equity" comments.
There is a moderate (0.66 on a scale of -1 to +1 using Spearman's rank correlation) correlation between timely review and overall satisfaction with code review. This question had the strongest correlation with overall satisfaction with code review.
Raw Data
edit| Question | Strongly Agree | Agree | Neutral | Disagree | Strong Disagree | Responses |
|---|---|---|---|---|---|---|
| I get timely review of my patchsets | 6 | 38 | 16 | 10 | 6 | 76 |
| I have expertise using Wikimedia's code review tools | 34 | 33 | 7 | 1 | 1 | 76 |
| I know how to find people to review my code | 21 | 35 | 11 | 6 | 5 | 78 |
| Our code review culture is healthy | 11 | 27 | 23 | 9 | 7 | 77 |
| It's easy to find and browse our code | 7 | 18 | 19 | 22 | 12 | 78 |
| It's easy to create a new repository | 5 | 11 | 26 | 16 | 13 | 71 |
| It's easy to change code without knowing the original author | 17 | 22 | 22 | 13 | 1 | 75 |