RFR: Section on sponsoring [v9]

[Magnus Ihse Bursie]

On Fri, 11 Aug 2023 01:38:01 GMT, David Holmes <dholmes at openjdk.org> wrote:

>> Jesper Wilhelmsson has updated the pull request incrementally with one additional commit since the last revision:
>> 
>>   Be mindful
>
> src/guide/reviewing-and-sponsoring-a-change.md line 29:
> 
>> 27: ### Conversation
>> 28: 
>> 29: The Conversation section is where all discussion around the PR happens. Note that this section is intended for technical discussion around the actual implementation of a change. This is not the place to discuss the appropriateness of the change or have overarching design discussions, that should happen on the appropriate [mailing lists].
> 
> I have to disagree about the last part - these days because the PR comments go to the mailing list and mailing-list-only comments (in a PR thread) go to the PR, it is indeed the case that design discussions and questions of appropriateness appear in the PR. Personally I prefer JBS to capture such things but the PR has become the one-place-shop in many cases and JBS commentary is typically very limited. Also JBS can't be used by a developer who needs a sponsor and doesn't yet have Author role.

I think PRs are typically used to discuss the *implementation* once you have one. If someone opens a JBS issue with e.g. a new feature, but with no corresponding code, then the discussion will happen there. 

>From time to time a PR is published with a concrete implementation, but then a discussion arises about the approriateness of the design, or even if the problem should be resolved at all etc. I don't think it is ideal if that kind of discussions happen on a PR, but maybe it's not that bad either. It is mirrored on the mailing list, and the only real problem in that case is that maybe the subject is bad -- it looks like discussions about a particular implementation, but is maybe indeed a discussion about the future direction on some functionality.

So, I dunno. Maybe we can recommend that people start up a new thread on the mailing list, if it turns out that some fundamental aspect of the PR needs to be discussed? I don't think it is a bad advice, but it could be formulated more softly as a suggestion.

-------------

PR Review Comment: https://git.openjdk.org/guide/pull/97#discussion_r1416980063