On 9/11/23 20:20, Toerless Eckert wrote:
On Sun, Sep 10, 2023 at 10:47:18PM -0400, Keith Moore wrote:
I don't like the idea of ADs delegating their responsibilities in general.
I was only thinking of a document-by-document basis. I've already seen more
than enough AD reviews deteriorate to directorate reviews recited by the responsible AD.
So, why should an AD not trust directorate members that have repeatedly shown to
provide good if not better reviews (on the specific topic) than the AD could do him/herself
to also drive the resolution of DISCUSS ? We also have chair delegates ;-)
The experience I had when I was AD was that when I asked directorate
members to do reviews, the reviews came from a wide variety of
perspectives. And if I got multiple reviews for a document (which
seemed only appropriate), the reviews, while sometimes providing
valuable insight, did not either reduce my AD workload or reduce my need
for expertise in that subject area. It was more work to review the
reviews than it was to review the document. So, based on my
experience, I am doubtful that "delegating" reviews really addresses the
problems that people claim it should. And I suspect it overall makes
IESG reviews less consistent.
(Different areas may have different experiences of the effectiveness of
directorate reviews. I was an Applications AD, and applications is a
very broad topic.)
I especially don't like that idea if ADs can use delegation to shift their
responsibilities to people for whom there are no remedies if they fail to
follow process or otherwise work against the community consensus or the
interests of the broader Internet community.
How would there not be the same remedies against such delegation activities.
It's of course possible to make remedies against such delegation
activities, but they do not exist at present and would require IETF
Consensus. And IETF is already overly bureaucratic, and this would
appear to make it worse.
Instead, it arguably becomes easier, because the AD as the administrative entity
is a first line of complaint, much easier to address than any redress against
an AD (which is a more convoluted process).
I don't think IETF management needs more ways to insulate
decision-makers from complaint. IMO IETF management is already too
insulated, is far too able to violate the IETF Consensus rules that have
been established for its operation.
Keith
