Closed bebatut closed 6 years ago
+/- 0 from my side. I don't think we have (currently) a problem that people do not review PRs. Moreover, assigning multiple persons to PRs can have a counter effect that no one feels responsible. I tend to trust people, that they feel responsible for a certain topic - if a PR is really open for a long time we can still manually ping people. My feeling is that this generates a lot of unneeded spam, without clear guidelines (how many assigned people need to agree?) and (currently) unneeded.
Either way, this file should be auto generated imho if we really need this.
I am with @bgruening. This repository does not seem to have a problem with reviewing PRs so this feels like premature optimization to me.
To be frank I think this repo could benefit from a bit more anarchy e.g. allowing trusted folks directly pushing.
-0
I dunno, I like it, not everybody watches this repo that closely maybe and they may appreciate the ping. And I feel like a lot of the reviewing is done by B&B rather than topic maintainers currently ..as long as people can opt out of being added here if they don't want to be? ..and I'm not sure it would make people feel less responsible, it could also make them feel more invited
I could even imagine people appreciating this on a tutorial-level ..if you contributed a tutorial you may like to be notified of proposed changes without having to watch the repo and see /all/ the activity/noise coming by. Maybe if we autogenerate this from the metadata files we can let people set their preferences in the contributors.yaml file orso (with whatever default we decide is best)
ping also @Slugger70 @Stortebecker @dpryan79 @joachimwolff as you're listed as topic maintainers but I don't seem to be able to request a review from you for some reason
@shiltemann they do not have write access (good limitation of codeowners to know)
@martenson ah yes, good to know indeed, maybe we should give topic maintainers write access anyway so they can merge PRs in the topics they're responsible for? The tutorial-level idea will be harder unless we go bioconda style and just add everybody (which may get you the anarchy you wanted ;)) but let's see if there is any demand for any of this first, I have no idea
Granting write to topic maintainers might be good. @joachimwolff asked me to look at #540 before we realised that neither he, nor @dpryan79 (the other methylseq maintainer), nor I (one of the admin stuff maintainers) had the ability to merge.
it could also make them feel more invited
Just speaking for myself, I don't feel invited if I get spammed from a bot. In contrast to be friendly pinged by a community member.
if you contributed a tutorial you may like to be notified of proposed changes
This is different from Maintainers, right? This PR will not solve this problem for Contributors.
repo and see /all/ the activity/noise coming by ...
I don't see where IUC or our other repos are different from this one here. And we never had this issue in others. This feels way over engineered and less human for me. But maybe I'm getting too conservative ;)
A simply rule how to name PRs and people can filter messages, all people not only maintainers.
Also labeling goes a long way.
@bgruening sure, fair enough.
Github notification just feel a bit all-or-nothing for me sometimes (e.g. I would love it if I could get notifications only for the mothur recipe in bioconda but not all the rest) and I do miss things in busy repos sometimes because of this, especially in periods when I'm a bit less active on GH (guess what I really want is more fine-grained control over github notifications)
..for me personally this PR doesn't add much as I already keep a close eye on the repo, and if all of the maintainers feel the same we are trying to solving a non-existent problem, agreed
This is different from Maintainers, right? This PR will not solve this problem for Contributors.
oh sure, this was just me thinking we could add tutorial-level rules here for contributors (if they so desired), but that is probably overengineering it, yes (and before I realized this only works for people with write access)
@martenson good point about the labels
For me, it does not matter much if people are pinged automatically or personally. And yes, getting write access to update tutorial-related stuff more frequently would be nice.
I have invited @Slugger70 @Stortebecker @dpryan79 @joachimwolff as collaborators with write access to this repo.
Thank you very much for your contributions!
thanks @martenson! ..don't forget @erasche :)
@shiltemann does he deserve it? 🤔
(🤣)
jokes aside, @erasche has access through the committers group, he has maaaaaany repos to push to
@martenson Thanks much!
@martenson Merci!
@martenson are you sure? there is a grey checkmark above in front of @erasche's name indicating no write access at time of review (and confirmed by comment) ..looks like you don't get automatic write access from being committer?
@shiltemann I think you do, but in the meantime @bgruening restricted pushing to master
of this repo (so now almost nobody does), please talk to him
@martenson ? @erasche was simply not added. I added him now.
as of just now ...
(Apologies to everyone else who is getting constant emails from this that they don't care about)
@bgruening that is not true, you restricted master
pushing to yourself, B! and Saskia. I told you it affects PRs. ;)
@martenson look at the grey checkmark above
its green now in Erics case. Thats what I'm talking.
@bgruening that just means the CI did not report error, doesn't it?
edit: I see which checkmark you mean now, but they still cannot merge to master because of the protected status
Ooh! I have a merge button now. Thanks BB!&S <3
As suggested by @erasche in #489, this PR add a
CODEOWNERS
to have automatic PR review requests of the maintainers of each topicCan you check if it is ok with you?