[dev][all] note to non-core reviewers in all projects
marios at redhat.com
Tue May 21 14:26:01 UTC 2019
On Tue, May 21, 2019 at 4:34 PM Brian Rosmaita <rosmaita.fossdev at gmail.com>
> Hello everyone,
> A recent spate of +1 reviews with no comments on patches has put me into
> grumpy-old-man mode.
> A +1 with no comments is completely useless (unless you have a review on
> a previous patch set with comments that have been addressed by the
> author). I already know you're a smart person (you figured out how to
> sign the CLA and navigate gerrit -- lots of people can't or won't do
> that), but all your non-comment +1 tells me is that you are in favor of
> the patch. That doesn't give me any information, because I already know
> that the author is in favor of the patch, so that makes two of you out
> of about 1,168 reviewers. That's not exactly a groundswell of support.
> When you post your +1, please leave a comment explaining why you
> approve, or at least what in particular you looked at in the patch that
> gave you a favorable impression. This whole open source community thing
> is a collaborative effort, so please collaborate! You comment does not
> have to be profound. Even just saying that you checked that the release
> note or docs on the patch rendered correctly in HTML is very helpful.
> The same thing goes for leaving a -1 on a patch. Don't just drop a -1
> bomb with no explanation. The kind of review that will put you on track
> for becoming core in a project is what johnthetubaguy calls a
> "thoughtful -1", that is, a negative review that clearly explains what
> the problem is and points the author in a good direction to fix it.
whilst i agree on all you wrote, imo a -1 with no comments is worse than
a +1 with no comments. If you dislike my patch enough to -1 it tell me what
i need to change in order to fix and get your vote
> That's all I have to say. I now return to my normal sunny disposition.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the openstack-discuss