Otto Kekäläinen <o...@debian.org> writes:
> For most people it is surely accidental, and sending a reminder to
> check for open MRs like I did today can help.
You didn't send a reminder. You sent a hostile message accusing anyone in
Debian who uploaded a package without looking at MRs first of
intentionally ignoring MRs and ruining the experience of contributing to
Debian. A reminder would be something like:
I've recently run into a few cases where a new contributor to Debian
created an MR on Salsa which received no response, and then saw the
maintainer later upload the package with equivalent changes while
having apparently ignored or not seen the MR. I know this is often
unintentional, but it can feel rather demoralizing.
Could maintainers who want to accept MRs for their packages try to
remember to check Salsa when working on a package and see if there are
any pending MRs to merge? If you don't want to accept MRs, please
disable them in Salsa so that people don't invest time and effort in
unwanted MRs.
I feel bad when I do this, and I appreciate the reminder! I don't
appreciate being accused of intentionally ignoring MRs just to make people
feel bad, since that's not what I'm doing.
> I am contemplating if I should share the cases of bad behavior my
> mentees experienced when trying to contribute, as you don't seem to
> believe that such could have happened.
That is not even remotely what I said. Come on now.
> For now I think that just reminding people about checking MRs and
> reviewing them as a general advice is the best course of action.
I think that's a fine idea, and I would welcome actual reminders rather
than accusations. Even better is to make that information more visible in
places where I already look so that I don't need a reminder.
> Also note that I suggest people to *review* MRs, I didn't say at any
> point that one must accept and merge all contributions.
Again, just to set expectations, I by and large am not going to do a back
and forth with a contributor about an MR. I simply do not have the time. I
am either going to merge it and fix it if needed, or I am going to close
it with some explanation and either make a different fix or no fix.
Reviewing an MR and explaining to someone how to fix it the way that I
would like it is a lot of work and not fun, so I save that energy for my
day job where it's part of the job.
I realize people learn better from interactive reviews, but I am not at
present volunteering as a teacher.
--
Russ Allbery (r...@debian.org) <https://www.eyrie.org/~eagle/>
