On 2024-03-18, at 04:56, Brian E Carpenter <brian.e.carpenter@xxxxxxxxx> wrote: > > You said (the context is below): > >> Doesn't matter what's said on the list, the document in the tracker is the source of truth; > > I don't see why it isn't equally true that: > > "Doesn't matter what's said on github, the document in the tracker is the source of truth" > > because that's a matter of definition - the most recently published draft *is* the draft. github is just a manifestation of a formal or de facto design team. We've had design teams and off-list discussions for decades. This. I’ve sometimes run into people who think we first need to establish working group consensus to merge a PR (*). WG chairs are of course free to set up rules like that, but then I think we don’t even need working group consensus to submit an updated I-D — that is exactly done so we have an efficient way to check whether the working group agrees with the direction proposed by the I-D authors. But again, please talk to your WG chairs about how your WG handles this; e.g., the WG may have an “implementation draft” label that it only gives to I-Ds after a mini-WGLC (which probably will allow open issues), with more sketchy drafts interspersed. Grüße, Carsten (*): It is not particularly bright to merge a PR that has significant opposition, but here “rough consensus” may apply. If you don’t like what your authors are doing, ask the chairs to find better ones, it is in their remit to make such decisions.
