--On fredag, januar 21, 2005 12:14:32 -0500 John C Klensin
<john-ietf(_at_)jck(_dot_)com> wrote:
--On Friday, 21 January, 2005 10:44 +0100 Harald Tveit
Alvestrand <harald(_at_)alvestrand(_dot_)no> wrote:
As you will probably be totally unsurprised to see, I'm
pushing back the prospective dates for IASA BCP approval again.
I think we are "close, but not finished" with the various
clarifications and process polishing - and still don't have a
workable consensus on the appeals issue.
...
On my previous timeline, January 12, I promised to ask the
questions relative to the -04 draft:
- Is the draft now good enough?
- Do we need to reissue, lengthen or restart the Last Call?
I think the first question got answered with a "no" - hence
this plan, and the new revisions.
Does anyone wish to argue that the second question need to be
answered "yes", or that there are other things that make this
timetable an unreasonable one?
Harald,
I'm not prepared to argue a personal need for another
several-week Last Call, but want to make two observations:
(1) When one looks at the number of changes that have been made
since the initial Last Call, and the number of those that have
actually been substantive, I think a narrow reading of our
procedures _requires_ such a Last Call and that proceeding
without it would be easily appeal-able. From that perspective,
what you are asking for is broad community consent to not hold
that Last Call. That is perfectly reasonable, but we should all
understand it in the same way.
I understand it - we could argue about whether the procedures were written
with the intent that such a narrow reading entails, but that's a long and
separate discussion. I do think we have to have broad community consent (as
represented by the readership of the IETF list) to not hold a second Last
Call.
(2) While I don't think we need four weeks, I'm concerned about
having only three working days (between the announcement
sometime on Friday 28 Jan to the IESG approval call on 3
February) to review the final document. While I think the
editors have done an astonishingly good job of pulling all of
this together, it is a lot of work and we know that a few things
that were agreed to slipped through the cracks of -04. So I
want all of us to be able to study -06 or whatever the IESG is
going to vote on. Three working days is not enough -- if I
recall, it even violates the IESG's internal minimum for WG
document availability ahead of a decision. My recommendation is
that:
* You not try to take a formal IESG decision until
Monday, 7 February.
I'll make that "seven days after the last version of the I-D is published".
Makes sense to me.
However, sooner or later we've got to stop fixing the small bugs - at the
time when the "presumed to be final" I-D is published, I will have to ask
people to only raise new issues that they feel are "show-stoppers" - this
is the same discipline that we've tried to enforce in the IESG with regard
to second and subsequent reviews. (Note: According to Friday's schedule,
-05 is not "presumed to be final".)
* Because this is not a standards action that is clearly within
the IESG's responsibility, and because the foundation documents
that got us here were joint IESG/IAB responsibilities, the IAB
should be polled on its opinion of the community consensus. The
information from that poll should be supplied to the IESG well
in advance of that Monday decision window and, if there are
dissent(s), the dissenter(s) should be encouraged to post their
concerns to the IETF list.
Reasonable thing to do.
* If the IESG is unanimous on the decision, you should
be able to take that decision by email and to do so quickly.
* If the IESG is not unanimous, then the concerns of the
dissenting IESG members should be posted to the IETF
list for comments and further input (and a period of at
least a could of days) before the IESG meets (presumably
by phone) to determine if it can reach consensus under
its ordinary voting rules. I am assuming that, if doing
this quickly is important enough to justify any
accelerated handling (i.e., not a four-week Last Call on
-06), then the IESG can find time for an extra
teleconference if that is really needed (and,
conversely, that the desire to avoid such a call may
motivate people to reach consensus :-( ).
We discussed this in the IESG telechat on Thursday last - the IESG members
agreed that they would raise DISCUSS-sized issues on the list, not via the
balloting mechanism. So given that these can be resolved, I believe there
should be few problems in achieving IESG consensus.
If there is really adequate community consensus behind this, and
the IAB poll and IESG action reflect that consensus, this
"costs" only a weekend and will, IMO, leave us all feeling much
more secure about the result. If there is still significant
dissent --either in the community or within the IAB and/or IESG
-- then airing and resolving it by this sort of pre-approval
mechanism is much faster than dealing with a possible appeal.
Thanks!
Harald
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf