Improving Merge Request review processes

Tue Jun 27 12:01:26 UTC 2023

One common technique is to put all the MR instructions gobbedlygook into a
, so they don't end up in the actual description. I
kind of agree the gobbeldy gook should be on the wiki, though, with a
simple  in the
actual template itself. But even that kind of falls down because the link
wouldn't actually be clickable. Maybe a big comment is fine.

At any rate, +1 to making the template an actual template.

On Tue, 27 Jun 2023 at 12:35, Simon Peyton Jones <
simon.peytonjones at gmail.com> wrote:

> The majority of contributors don't have the rights to modify labels on
>> merge requests. It might be good to add a few words of advice for that
>> group: What should *they* do if their MR needs attention?
>>
>
> Oh!  That is bad.  It's no good us saying "add label X" if they can't add
> a label.  Can we give everyone rights to modify labels?
>
> Also: I have noticed that many MRs (hilariously including !10709) start
> with "Thank you for your contribution to GHC".  That always irritates me:
> it suggests that the author has not really bothered to write a proper MR
> description.   The template isn't really a template: it's more *guidance *about
> how to submit a MR, which we also supply at
> https://gitlab.haskell.org/ghc/ghc/-/wikis/Contributing-a-Patch.    (And
> if the template is widely used, as it seems to be though I don't know how
> people find it initially, it should point to that wiki page.)
>
> I'm not quite sure how to resolve this.  Maybe make the template be truly
> a template, and point to the wiki page for all content?
>
> Simon
>
> On Tue, 27 Jun 2023 at 06:53, Bryan Richter <bryan at haskell.foundation>
> wrote:
>
>> The majority of contributors don't have the rights to modify labels on
>> merge requests. It might be good to add a few words of advice for that
>> group: What should *they* do if their MR needs attention?
>>
>> Thanks for this effort! The vast, vast majority of GHC contributors are
>> the ones who haven't even heard of it yet, or aren't even born yet. High
>> quality information on how things work and how to get started will be the
>> key to growing the ranks of GHC experts over time.
>>
>> On Fri, 23 Jun 2023 at 18:40, Matthew Pickering <
>> matthewtpickering at gmail.com> wrote:
>>
>>> I have modified the MR template -
>>> https://gitlab.haskell.org/ghc/ghc/-/merge_requests/10709
>>>
>>> and updated the wiki page that simon points out.
>>>
>>> Matt
>>>
>>> On Fri, Jun 23, 2023 at 11:25 AM Simon Peyton Jones
>>> <simon.peytonjones at gmail.com> wrote:
>>> >
>>> > Terrific.
>>> >
>>> > We just need to add some guidance for contributors about when and how
>>> to use the new tag.
>>> >
>>> > Where should that guidance live?  We have the Contributing to GHC wiki
>>> page.  It in turn (not very prominently) points to Contributing a patch.
>>> Are these our primary pages?  If so, perhaps the latter is the one to edit?
>>> >
>>> > Simon
>>> >
>>> >
>>> >
>>> > On Fri, 23 Jun 2023 at 11:14, Matthew Pickering <
>>> matthewtpickering at gmail.com> wrote:
>>> >>
>>> >> We discussed this in the meeting on Tuesday.
>>> >>
>>> >> The conclusion was that
>>> >>
>>> >> * We now have a new label "Blocked on Review", which people can add to
>>> >> merge requests if they are blocked waiting for a review.
>>> >> * The "Reviewer Group" is taken to be the same as the "GHC Team" (see
>>> >> https://gitlab.haskell.org/ghc/ghc-hq/-/tree/main), hence people
>>> >> trusted to be part of the GHC team are expected to perform review as
>>> >> well.
>>> >> * We will take up some of the time on the Tuesday meeting by talking
>>> >> about merge requests which are blocked and assigning them to people
>>> >> for review.
>>> >>
>>> >> Cheers,
>>> >>
>>> >> Matt
>>> >>
>>> >> On Mon, Jun 19, 2023 at 10:03 PM Matthew Pickering
>>> >> <matthewtpickering at gmail.com> wrote:
>>> >> >
>>> >> > Hi all,
>>> >> >
>>> >> > Recently there has been some discussion about better systems and
>>> >> > processes for keeping the flow of merge requests going smoothly
>>> >> > through the review process. It has become clear that we need to be a
>>> >> > bit more deliberate in handling merge requests in order to make sure
>>> >> > we can correctly triage, review and merge the many fantastic
>>> >> > contributions we get to GHC on a daily basis.
>>> >> >
>>> >> > Therefore we are proposing to introduce a "GHC Reviewer Group" whose
>>> >> > members will share collective responsibility for ensuring that MRs
>>> >> > make their way smoothly from creation to merge.
>>> >> >
>>> >> > The description of the role and responsibility for this group can be
>>> >> > read and commented on here:
>>> >> >
>>> >> >
>>> https://docs.google.com/document/d/1FK9mryjC82DM6e5yxP7BOgu94As2bXccb55L8OrjPf4/edit?usp=sharing
>>> >> >
>>> >> > The motivation for this proposal is two-fold.
>>> >> >
>>> >> > * Ensuring that MRs are reviewed and triaged in a timely manner.
>>> >> > * Documenting where the responsibility for MR reviewing
>>> >> >
>>> >> > We welcome any discussion about this document and other ideas about
>>> >> > how to improve the systems in this regard.
>>> >> >
>>> >> > Cheers,
>>> >> >
>>> >> > Matt
>>> >> _______________________________________________
>>> >> ghc-devs mailing list
>>> >> ghc-devs at haskell.org
>>> >> http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs
>>> _______________________________________________
>>> ghc-devs mailing list
>>> ghc-devs at haskell.org
>>> http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/ghc-devs/attachments/20230627/8058068c/attachment.html>