[mpowr] Re: [Solutions] Further work on WG (chair) roles - MPOWR WG proposal

["James Kempf" <[e-mail]>]

Dave,

> It depends on the defined task.  If the task is "coordination", then the
> responsibility for tracking document state is entirely appropriate.  It
> does not need accompanying "authority".
>
> Let me suggest that "authority" is looking more and more like exactly
> the wrong construct for fixing IETF problems. Focusing on authority
> ensures that we will need to have curbs against abuse, because there are
> certain to be some abuses.
>
> What is needed is facilitating collaboration. This means that
> "authority" rests with collections of people who reach agreements. Any
> individual who appears to have authority has it by virtue of delegation
> from a collection of participants, for the purpose of efficiency.
>

I think it does involve authority. 2026 only allows the WG chair to judge
rough concensus. If there is rough concensus then the chair must forward the
document to the IESG, regardless of how broken (s)he may think it is. Given
the current rules in 2026, the WG chair can't simply hold the document for
review and send it to a third party group of reviewers, like SIRS, then come
back and insist that the WG fix the problems. (S)he can only do that if the
reviewing party is the IESG. And the chair can't even send along a list of
problems (s)he thinks need attention and request that the review group pay
particular attention to, even if it is the IESG. The chair is expected to
represent the WG concensus to the IESG, nothing more.

We're facing this problem right now in Seamoby. We sent one document out to
a review committee at my request (I'm the co-chair), and it came back with a
list of problems the reviewers wanted to see fixed. But the reviewer's
comments were not any more binding on the design than any others, and the WG
members quibbled with the reviewers on the WG list until the reviewers gave
up, essentially a successful DoS attack.

This isn't "co-ordination", because the WG chair is making a judgement about
the need for enforcable QA on the spec, just as a manager would when doing
software development. If you don't like the word "authority" then pick
another one.

> Let's take the specific case of a rogue working group and a
> knowlegeable, diligent chair. How is the chair -- or the AD -- to
> "control" the wayward choices of a misguided working group? Surely the
> superior expertise of the chair should prevail?
>
> The problems with assuming that the chair has that expertise, and then
> assuming that it will always be applied properly. We are not suddenly
> going to select better Chairs. Some have enough of the relevant types of
> expertise and some do not. Some will invoke it wisely and some will not.
>
>      Let me suggest, instead, that the core test mechanism in the IETF
>      is the ability to recruit others to a point of view.
>
> Hence, the expertise of the Chair (or anyone else, for that matter)
> should be used to recruit a collection of _other_ participants to
> counter the misguided views of the working group.
>
> One might even think that independent, senior IETF reviewers might be
> helpful to this goal...
>

See comments above. Unless the review makes enforcable changes, it's not
worth anything. The WG can continue to  quibble until the reviewers give up.
There needs to be some check on WGs that don't deliver quality. If you don't
like the WG chair exercising their design judgement enforcably, then they at
least need to be able to decide what they don't like and submit it, along
with the document, to some third party group that has the power (sorry for
swearing) to enforce quality.


            jak


_______________________________________________
mpowr mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/mpowr