Re: Appeal to the IAB on the site-local issue

[Scott Bradner <sob@harvard.edu>]

In private email I was asked to review the "tape" of the IPv6 discussion in
San Francisco.

ftp://limestone.uoregon.edu/pub/videolab/video/ietf56/ietf56%20-%2003202003%20-%20INT%20ipv6.rm

The SL discussion starts at 1:02 into the session.

The chairs first presented a set of slides talking about various SL related
options. The presentation was interrupted with questions from the floor
quite a few times during the discussion of the "exclusive" model (wherein a
v6 node would be a SL node of a global addressing node) - it was clear to
me that there was quite a bit of confusion about this model.  There were
few interruptions during the discussions about the other models.

The chairs opened the floor for general discussion at 1:39 into the
session.  The discussion was careful and extensive.  After a while it
became clear, as noted by Thomas, that more people were arguing for
eliminating SL than had been the case in the past and few people were
arguing for SL addressing. Those that were arguing in favor of SL mostly
said that SL and v6 NATs were going to happen anyway but no one seemed all
that concerned that the IETF define such addresses (e.g. Deno pointed out
that people would just pick their own if the IETF did not.)

At 2:07 into the session the chairs conferred and said that they would ask
a simple yes or no question (in reality they asked two questions) about
deprecating IPv6 SL addresses.  (Not eliminate them in that the sense that
the prefix would not be reassigned for other uses.)  Margaret noted that
the simple questions covered a lot of details that were not called out.

After 10 minutes of discussion to clarify the intent of the questions
Margaret asked for a show of hands for:
	1/ how many people want to deprecate the use of IPv6 SL addresses?
	2/ how many people do not want to deprecate the use of IPv6 SL
	  addresses?

She asked the first question twice so they could get a count of hands the
second time.  The result was 102 hands in favor of deprecating and 20
opposed.  The chairs declared that there was rough consensus in the session
to deprecate the use of IPv6 SL addressing but that this consensus would
now be taken to the list to verify.

After my review of the tape I have come to the following conclusions:

	1/ there was a lot of confusion over the details of at least the
	   exclusive model for the use of SL

	2/ it was a clear sense of the people who chose to express their
	   opinion in favor of deprecating IPv6 addresses (I have no 
	   way of knowing what fraction of the room chose to express 
	   an opinion because the IETF web site says that no meeting 
	   report was received so I have no way to know how many people 
	   were in the room.)

RFC 2418 says:
   ... Decisions reached during a face-to-face meeting about
   topics or issues which have not been discussed on the mailing list,
   or are significantly different from previously arrived mailing list 
   consensus MUST be reviewed on the mailing list.

   and

   In the case where a consensus which has been reached during a face-
   to-face meeting is being verified on a mailing list the people who
   were in the meeting and expressed agreement must be taken into
   account.  If there were 100 people in a meeting and only a few people
   on the mailing list disagree with the consensus of the meeting then
   the consensus should be seen as being verified. 

I followed the discussion on the mailing list.  In my opinion, there was
considerable less clarity of view on the list than in the meeting.

I know of no way to figure out what overlap there was between the people
expressing their opinions on the list and those who expressed their opinion
during the face-to-face session but it seems likely that the ratio of
opinion is somewhere between 3/4 (the ratio reported by Margaret about the
mailing list discussion) and 4/5 (the ration in the meeting).  

While I would rather see rough consensus not wind up with 20 or more
percent of the IETF community (or at least the part of the community that
expressed an opinion) disagreeing with a decision. (That many people may
have input that the rest of the community dismisses at its peril.) I have
to support the working group chair's assertion that rough consensus was
found in this case.

I will follow up this note with two others later today but I wanted to
limit this note to the specific topic of the level of consensus and
confusion.

Scott