Paul Moore schrieb am 15.05.19 um 17:18:
I don't really follow this sort of PR discussion, as the github emails don't tend to have sufficient context on what's being said
I agree, although there is also an upside to it. PR discussions can be more easily constrained to reflect the exact reasoning behind the specific change, whereas more general PEP discussions, especially in mailing list threads, are more likely to cover broader (sets of) topics and/or get distracted and jump between topics. So that's an improvement, I think.
Not every PEP change is easy to discuss as a PR, though.
multiple fragmented forums for discussion. It feels a lot harder these days to keep track of all the discussions/decisions going on.
+1
Discussions easily get out of the scope of a PEP PR or the original topic, which makes it impossible to know when something relevant happens to get discussed in one of a dozen places that can be used to discuss them.
E-mail threads obviously have the same problem, but at least they are still part of the same mailing list, so subject changes are relatively easy to detect when … the subject changes.
Same for conferences, they are great for discussing complex topics and working together to improve the understanding of a matter, but then someone has to sit down and write up the outcomes so that the general discussion can start (or continue) in the public places.
I would prefer a sort of an "open first" principle, where things are discussed on python-dev (Python) or python-committers (processes), unless there is a reason not to (such as PRs, SIGs, voting, …). That gives everyone a good handle to stop a discussion and say "let's move this to place X".
More generally, there needs to be a simple scheme or checklist that makes it easy to detect when a discussion should be moved elsewhere, and preferably to which place exactly. At least for the 80% case.
Stefan