Hi John,
I agree with you, it should not be allowed, but I think it is important also the session's chairs/ADs must have a say in such special cases there may be a need for a presentation or a document that is needed for the session/WG/Charter. There is always a possibility that presenters don't present for some reason (causes change of schedule/agenda at the start of the meeting to be agreed_on/discussed), so the chair is responsible to manage the session to be useful the most to attendance..
For my understanding the deadline (for conferences and meetings) is to give time for participants read the documents and also for the author/presenter to get enough/fair time to present, but also if there is some more time available/open_mic at the end of meeting, then any one can present any thing which is needed/maybe_needed by participants.
Best Regards,
AB
On Fri, Mar 15, 2024 at 9:27 PM John C Klensin <john-ietf@xxxxxxx> wrote:
Hi.
For many, many, years we have had a cutoff for posting I-Ds
about two weeks before an IETF meeting starts. That cutoff was
established to give people time to read documents (especially
because, predictably, many documents would be posted just before
it), figure out which WG meetings they needed to attend, prepare
comments, etc. We are now seeing pull requests that alter
substantive parts of documents posted on Github within a short
time before IETF starts and during the window when new or
revised I-Ds are not allowed without special circumstances and
specific permission from ADs.
Allowing that seems to contradict, or at least seriously weaken,
the principle of having documents available well in advance of
meetings. If they are announced only to the mailing list of the
relevant WG(s) (and sometimes not even that widely), it seems to
me that they impede both WG meeting discussions in which
everyone has the same starting points and openness to IETF
participants who have not signed up for the WG mailing list
(newcomers included).
Is the IESG considering some guidance on this subject or is it
considered unnecessary?
As an almost-separate question, if the "real" version of an I-D
that is expected to be discussed in meetings and on mailing list
is the the draft plus the cumulative effect of pull requests
(some by other than the listed authors), should that be more
clear and reflected in the datatracker?
thanks,
john
