I'm aware of the tools team proposal. But I claim it
illustrates the problem. See below.
--On Tuesday, 19 October, 2004 01:03 +0200 Henrik Levkowetz
It seems to me that this is one of the reasons why discussion
of these proposals/plans with the community is important. It
is not just a matter of approval of a new rule (although that
is important), but the fact that the community can often come
up with clever solutions that the Secretariat, or IESG, on
their own, might not discover. I don't know if it would be
worth the trouble, but "either get WG chair pre-approval a
week in advance _or_ the WG Chair must submit the document"
would seem to me to be a much more reasonable rule than the
current one, which encourages individual-submission naming,
followed by reissuing of an identical document under the WG
name, which makes documents harder to track.
I don't have any input on exactly how, originated by whom, the
current deadlines have come to be.
However, if we're starting to discuss the mechanics of what
be done to move the deadlines closer to the meetings again,
I'd like to point out that the ietf tools team, as it's first
task, has been formulating requirements for a tool to automate
submission, so that secretariat workload can be eliminated
or severely reduced as a factor in draft posting deadlines (and
WG chair approval deadlines).
But you see, the secretariat workload has already, somehow, been
reduced to the point that we are clearing the queue well in
advance of the meetings, unlike a few years ago, when the
deadlines were set and we routinely went right up against the
meeting. If your "reduce the load enough that things can be
gotten out faster will result in deadlines closer to the
meetings" hypothesis is correct, then I'd expect that we would
already have had a review --initiated by either by the IESG or
the Secretariat and discussed with the community-- about how
much closer the deadlines could be moved, followed by
implementation. Instead, we have acquired a new rule that
pushes the deadline even further out.
I know that this is a little extreme, but, based on that
experience, it is equally reasonable to assume that, if fewer
cycles are required to process I-Ds before an IETF meeting,
someone will wake up and, without consulting the community about
priorities, decide it is useful to impose several _more_ process
steps, since there would then be time for them within the
We believe that the requirements are close to done, and hope
the tool can be produced and deployed fairly soon. This
should make the draft posting deadlines a matter of deciding
is optimal for the community. The tool should make both
and chair approval possible as close to the meeting dates as is
found to be desirable.
Sure. But assuming that either a careful review of what is
"desirable" or any movement at all, will happen is, however
rational, not supported by recent facts or experience. The
creation of the tools is really independent from setting of the
Ietf mailing list