I just took a quick (~1.5 minutes) look through all of the #Wikimedia-Git-or-Gerrit tasks and it looks like a rename to Gerrit would be sensible based on the content/purpose of the tickets. I take responsibility for re-triaging any tickets that are oddballs, if any.

#Wikimedia-Git-or-Gerrit is used as a catch all for anything that is git related. That includes among other:

• git support questions
• Gerrit
• gitblit/git.wikimedia.org
• git-review
• random tools using Gerrit API

I would rather triage Gerrit issue from #wikimedia-git-or-gerrit to a new Gerrit project. Then most probably mass decline / dispatch all the rest. Once there is no open bug we can archive the project, bugs that are closed would be left with it but that is good enough imho.

Or of course, rename it Gerrit and create another project #git-related to move stuff to :D

I created a "really gerrit" on the dashboard https://phabricator.wikimedia.org/project/sprint/board/330/ so we can potentially move true Gerrit issue there.

In T112039#1628172, @hashar wrote:

#Wikimedia-Git-or-Gerrit is used as a catch all for anything that is git related. That includes among other:

• git support questions

That can stay there, but really aren't great for a task management system. It's not for support.

• Gerrit

would stay, obviously

• gitblit/git.wikimedia.org

going away, and/or we could make a Gitblit or #gitwikimediaorg project for them

• git-review

would stay, since it's 'a part of' gerrit

• random tools using Gerrit API

should be in their own project, unless the issue is with the gerrit api itself

I would rather triage Gerrit issue from #wikimedia-git-or-gerrit to a new Gerrit project. Then most probably mass decline / dispatch all the rest. Once there is no open bug we can archive the project, bugs that are closed would be left with it but that is good enough imho.

That sounds like extra work without a benefit to me.

Or of course, rename it Gerrit and create another project #git-related to move stuff to :D

:)

In T112039#1628181, @hashar wrote:

I created a "really gerrit" on the dashboard https://phabricator.wikimedia.org/project/sprint/board/330/ so we can potentially move true Gerrit issue there.

#Projects are cheap and help organize, workboards take more effort to manage and are kind of hidden and inconsistent.

I still strongly vote for renaming to Gerrit and use it only for gerrit (the vast majority are about gerrit) and moving others to their rightful place.

I've now gone through more thoroughly, out of the 114 tasks #Wikimedia-Git-or-Gerrit only 7 (seven) are not issues about Gerrit.

About Github:

• T109939: For mirrored GitHub repositories, actually give the canonical source Gerrit URL in the repo description

About Gitblit:

• T94320: Improve monitoring of https://git.wikimedia.org/
• T83702: git.wikimedia.org is unstable
• T59340: Purge git.wikimedia.org URLs from Varnish by subscribing to gerrit stream-events
• T58953: Gitblit backlink to mediawiki.org/wiki/Special:Code for old commits should be https

About Gitblit and Gerrit together (so would live in both #/gitblit and #/gerrit:

• T56741: Better cross-reference git.wikimedia.org and gerrit.wikimedia.org

About Phabricator/Gerrit integration:

• T48983: When a new extension repo is created, a new Phabricator project should be created

I'd keep things like "T90900: Remove obvious mistakenly added refs in mediawiki/core repo" in #/Gerrit for now as that is the place we'd have to do the work.

I think this change would be pretty minimally invasive (as described) and I'd like to do it by the middle of next week.

In T112039#1630795, @greg wrote:

this change would be pretty minimally invasive (as described) and I'd like to do it by the middle of next week.

Appreciated! And big thanks (or rather my deep respect) for wading through all those tickets.

In T112039#1654795, @greg wrote:

• Wikimedia-GitHub (for that one task that was the oddball out, there might be more that are resolved, I haven't yet searched for those)

Re searching over closed tasks: I deem this not worth my time after doing a quick look, it'd take too much reading of the actual tasks instead of just the titles. Dwell not in the past, but look into the future.

Well done @greg thank you very much :-}

In T112039#1654810, @greg wrote:

Re searching over closed tasks: I deem this not worth my time after doing a quick look, it'd take too much reading of the actual tasks instead of just the titles. Dwell not in the past, but look into the future.

I agree, not worth anyone time and that results in a spam of notifications to all subscribers on tasks that have been closed a long time ago.

It is not like we deal with medical records or money :-]