Re: Reducing the number of open issues

[Jean Abou Samra]

Le 17/03/2021 à 20:10, Jonas Hahnfeld a écrit :

> Hi all,
>
> there are currently 1066 open issues at LilyPond's GitLab repository:
> https://gitlab.com/lilypond/lilypond/-/issues
>
> I'd like to reduce this number: In the short term, I propose to close
> issues with ~Patch::abandoned (69 open issues) and ~Patch::needs_work
> (73 open issues) labels that come from patch tracking with SourceForge.
> They will stay around (including their links to Rietveld, as long as it
> stays alive), but not artificially increase the number of "issues".
> The caveat is that I remember some real problem reports with proposed
> patches, that were abandoned during review. I'll keep those issues open
> because they might need fixing either way.
>
> Any objections to this plan?
Agreed, I think it makes no sense to keep these issues
open. I would guess that in nearly all cases, it's
better to start fresh patches. If you're willing to
put in the effort to start new issues when the old
patch-tracking ones indicate a bug, I'm all for it.

> In the longer term, I think it could make sense to hold a "bug bash
> weekend" and try to pick some low-hanging fruits. That would not be
> limited to developers to write patches, but everybody to try and
> reproduce old reports with the then-current unstable release and close
> issues that are no longer relevant. Let me know if this sounds
> interesting and you'd be prepared to participate.

This message made me feel guilty about my
promised fix for #6093 which got lost in
other concerns... (Patch now at
https://gitlab.com/lilypond/lilypond/-/merge_requests/690.)

At any rate: this does sound sensible to me. I see
a lot of this low-hanging fruit in the tracker;
here are some examples that come to mind:

https://gitlab.com/lilypond/lilypond/-/issues/3345
https://gitlab.com/lilypond/lilypond/-/issues/6076
https://gitlab.com/lilypond/lilypond/-/issues/6086

There are also platform-specific issues, where we
might get information from users being able to
reproduce the problem or not in different configurations
I'm thinking of
https://gitlab.com/lilypond/lilypond/-/issues/6101
here.

I currently have extremely limited time; this should
get better by July or something like that. I'd be
willing to participate in such an initiative then.

Thanks,
Jean