On 17-Jul-23 07:26, John C Klensin wrote:
--On Sunday, July 16, 2023 16:07 +1200 Brian E Carpenter
<brian.e.carpenter@xxxxxxxxx> wrote:
On 16-Jul-23 15:50, Phillip Hallam-Baker wrote:
Why is attending biweekly con calls exclusionary?
It depends where one lives and whether one has a full-time
occupation. But really the point isn't there, it's whether the
result of such calls (which de facto are design team meetings,
not WG sessions) is brought back to a genuine WG plenary,
electronically or in person, for an effective debate.
I'd probably be more comfortable if these meetings were
pitched as open design team meetings; that would make things
clearer.
Brian,
I probably would too, except that "open design team meeting"
seems like a phrase that is likely to mean enough different
things to enough different people (as much of the discussion
subsequent to your posting may demonstrate) that it sets us on a
path toward needing to do a comprehensive procedural overhaul.
While I think it may be about time for that, experience
(starting with NEWTRK and continuing) suggests that starting
down that path does not lead to useful and substantive results
in finite time.
While I haven't thought about it enough to know if I agree, the
model phb outlines would almost certainly require such a
procedural overhaul even though one aspect of it would actually
take us a step toward where we were 30-odd years ago. I would
hope that no one would try to institute it by IESG Statement.
Noting Carsten's comments in response to Abdussalam and Keith,
it is likely that the use of Github should also point us toward
a procedural overhaul, the existence of RFCs 8874 and 8875
either notwithstanding. Or it might be that their publication as
Informational rather than as updates to RFc 2418 are
symptomatic. Still, I hope we can keep those issues separate
from issues about Interim meetings and the processes/
requirements to approve and hold them.
Noting the (IMO useful) exchange between Joel and Keith about
design teams, I believe one of our assumptions about them is
that they are focused on a narrow range of topics and relatively
short-lived. If a design team needs to meet regularly for
months (or probably even weeks) and/overs a large fraction of
the topics within the scope of the WG charter, it is either:
(i) Indistinguishable from the WG itself except for fewer
requirements for documentation and transparency.
(ii) A sign that the WG was chartered prematurely and hence
should be shut down until people are ready to propose a new
charter and to get work done according to WG rules.
I think there's a 3rd possibility, something like
(iii) The WG's topics include a lot of relatively minor
technical choices that need to be made but are non-controversial,
so most people in the WG don't care and will accept whatever
a subset proposes.
It seems to me that the first is a singularly bad idea and the
second, while appropriate, would be something for which the
community has no willingness to move on (or the IETF doesn't,
probably in part because they see no hope for community
support). As an exercise, think about how many WGs we have seen
shut down in the last decade because they were not actively
progressing toward the goals of their charters, emitting quality
documents into IETF Last Call, etc.
Or did you have something else in mind?
Not really. I just think it would be clearer what is really going
on if we label discussion by a subset as such. As has been
pointed out, it doesn't really change the rules, except that
"minutes" are replaced by bringing the design results to the WG
as a whole.
Given the lamentable quality of most WG minutes, that hardly
matters.
Brian
