On Fri, Apr 17, 2026 at 10:16 AM Dmitry Dolgov <[email protected]> wrote: > Somewhat late to the party (wanted to write right away, got distracted > for a few moments, and boom -- it's already few months old), but do you > consider gathering statistics based on Reviewed-by? It might give a > bit of appreciation for reviewing, which is commonly considered as a > thankless job, and will be more direct than just measuring amount of > emails to hackers.
It's worth considering, but one problem is that reviews vary extremely widely in terms of how much effort goes into them and how much they help the patch along. Another is that not all helpful emails are, in a strict sense, reviews. A third is that Reviewed-by only counts things that ended up getting committed, while reviews of things that don't get committed are also quite important. I don't really think that number of hackers emails is a very good metric, especially in this era of increasing numbers of AI-generated emails, and I wish I had something better, but I'm also not convinced that counting Reviewed-by headers would be that thing. Of course, that's also not to say it's bad, just that I'm not necessarily excited about doing more work for a number that I don't completely believe in. -- Robert Haas EDB: http://www.enterprisedb.com
