Re: Comments on draft-iesg-charter-02.txt

[Margaret Wasserman <mrw@windriver.com>]

Hi Scott,

At 03:50 PM 3/10/2003 -0500, Scott  Bradner wrote:
> >  Is there any document that gives the responsible AD
> > this authority?
>
> common sense does (in my opinion)
>
> 1/ the AD understands (to some degree) what concerns the rest of the IESG
> will have so it makes sense to get these fixed before bringing
> it in front of the IESG
>
> 2/ the AD need to be able to explain/support the doc to the rest of
> the IESG - if they are not ready to do that there is not much reason to
> bring it forward
I think that we may be caught in a semantic distinction regarding
what it means for a document to be "in front of the IESG".  If
the IESG chooses to delegate document review to the responsible
AD, and/or to engage in a two-step review process, that's fine
with me.

But, IMO, a WG chair should be able to send a document that
represents WG consensus (in the opinion of the chair) to the
IESG for review, and expect that the document will be sent to
IETF last-call and will result in one of the 5 outcomes specified
in RFC 2418 within a reasonable time period.  That's what it says
in RFC 2418, but that's not always what happens.

[To be fair, I haven't had this problem with the documents
that I've sent to Bert on behalf of the entmib WG.  They've
always been processed in a timely manner, and the WG has
always been actively engaged in any IESG feedback.  Thanks,
Bert.]

> I do think that this review can be, and has been, over done and
> sometimes is a too much a reflection of the individual AD's biases
> than of ensuring clarity or good technology
>
> I'm not sure what can be done to avoid the extremes and yet preserve the
> logical area management role (which includes making sure that the WGs put out
> good documents) but I think that saying "no AD review" will do anything other
> than increase the workload on the IESG
I would not have a problem with a quick sanity check by the
responsible AD before the document goes to last-call, full
IESG review, etc.  That could be considered a check on whether
the WG chair has properly done his/her job, which might be
reasonable in the ADs capacity as manager of the WG chair.

But I do have a problem with going through a lengthy process
to perfect a document based on one ADs opinion, only to go
through another lengthy process when the document gets to the
full IESG.  Once the document is deemed free of major surface
flaws and worthy of in-depth review, the document should be
reviewed by the whole IESG in parallel.  The current
serialization wastes a lot of time, and doesn't gain anything.

I'd also like all documents that are submitted to the IESG to
result in one of the five possible outcomes in RFC 2418, in a
timely manner.  Sometimes, with the current informal process,
it can be hard to distinguish between the wording and editorial
suggestions of one AD, and real technical/clarity issues for
which the IESG would block the publication of a document.

BTW, I think that there are also serious problems with the
quality of documents that WGs are producing, and with the way
that WG chairs run WGs.  I'm trying to make improvements in my
own group, and I'm working with Thomas, Jeff Schiller and others
to try to make it clearer to WG chairs what we are expected to
do, and to support us in that effort...  Although this thread
has focused on what I consider to be problems in the IESG,
I certainly don't think that the IESG is the only source of
problems.

Margaret