General comment on draft-ietf-problem-statement-00.txt

[Eric Rescorla]

"Harrington, David" <dbh at enterasys.com> writes:
> It is up to the WG to make sure the right thing gets done, and it is
> my responsibility to contribute to that discussion, to keep the
> discussion focused, and to make sure the WG considers that the
> "right thing" will also be dependent on what other WGs are
> doing. The AD is a useful resource to provide some input on what
> other WGs are doing, but I also supplement the AD input by
> discussing directions with the chairs of the other WGs and with
> contributors to other WGs. The AD helps me to identify which other
> WG efforts I should be aware of, and the AD can help identify useful
> contributors to that effort that might be willing to join in our
> effort as well to help provide insight into aspects of the other
> WGs' efforts. I can then make sure "my" WG considers that other work
> that is being done. So the AD helps me help the WG determine the
> right thing to do.
This is what I'm reacting to. The AD doesn't "help the WG
determine the right thing to do". The AD is in a position to tell the
WG what to do--and ADs do so fairly regularly.

It seems to me that there are two competing models for how IETF WGs
should work, which, for lack of better terms I'll call "Parliament"
and "Team".

In the Parliamentary model, the WG is essentially a collection
of individuals who need to come to consensus. In order to do
so, they've adopted certain procedural rules. The chair is
responsible for enforcing those rules. As such, the chair is
an extremely powerful member of the WG but in practice is
just another member and their job is to defer to the will of
the group.

In the Team model, the WG is a team formed to do a job
under the supervision of the ADs. The chair is the AD's
representative and their job is to make sure that things
get done. This includes imposing their technical
judgement where they believe it's approprirate.


Now, from a descriptive standpoint, it's clear that we don't
live in either of these worlds. On the one hand, the chairs
don't really run the working group--I think your message makes
that pretty clear, actually. On the other hand, the ADs 
fairly routinely impose stuff on the WGs even though the
WGs may have consensus to the contrary.

>From a prescriptive standpoint, a lot of the complaining
I hear about document quality seems to me to be of the
form that the chairs aren't really exercising independent
technical judgement the way that the ADs do. 

-Ekr

-- 
[Eric Rescorla                                   ekr at rtfm.com]
                http://www.rtfm.com/