Joel,
Thanks for writing this. I have some detailed comments, but perhaps I should
first start with my own perception of the meeting.
I traveled to this meeting as a part my trip to attend RIPE a few days, and to
catch a few different people in the hallways on separate topics. One datapoint:
I would probably not have made the trip just for RIPE this time (although I
usually do travel to them), nor would I have attended just for the LIM itself.
I was quite surprised by the small attendance, what I heard from the
secretariat was 30+ people registered for the interim. From a pure financial
perspective this meeting did not seem to make a lot of sense for the IETF, 30 x
100$ for a lot of secretariat effort/travel, and presumably money for room fees
and other costs. Most previous interim meetings have operated on a
hosting/sponsoring model, or run in zero cost environments. I'd say that
anything under hundred persons is probably easier to deal with in that way than
attempt to use an official organization for it.
But money is really a side issue, so lets not focus on the above too much. What
matters if we can progress the IETF's work:
The various meeting had a big difference between them. The SIDR meeting was
clearly a close group of people with an intent to solve a design problem, going
over an issue after another issue, involving people who would not have traveled
to the IETF, and at least from my perspective the meeting was making quite a
lot of progress in some issues that had been divisive earlier. This reminded me
of some earlier SOFTWIRE meetings, for instance, where the designers and
implementors met and made a lot of progress. An excellent meeting!
FWIW, in the IETF I would not have had time to go to SIDR, and I would not have
traveled anywhere to go to a SIDR meeting. But I'm glad I went to this one
because it allowed me to learn more about this technology, and some of the
layer nine implications at the IAB force me to try to understand its state.
I did not attend OPSEC, but I attended V6OPS. All discussions were useful, but
topics seemed somewhat disconnected, and they did not cover the whole range of
usual V6OPS work. It was clear that the room did not have the full set of key
contributors in the WG. I had prepared comments on a recent topic in the
mailing list, but the relevant people were not there to receive my comments.
Nevertheless, the V6OPS meeting did have a fair number of the usual experts in
this topic, as well as a couple of extra people from the European routing
community. I was able to have a high-bandwidth discussion on several topics
with the people present; albeit that was in the hallways and after the meeting.
We started one new draft for V6OPS based on this interim, and another (yet to
be completed) work item for the HOMENET WG. In the latter case I would not have
had the necessary people present in the IETF, so for me this was quite positive.
It is hard to tell why there was no more attendance in V6OPS. Wrong time and
place? Or true lack of interest to invest in some of the topics that we are
just not seeing it in the IETF because everyone comes there anyway?
Conclusions? Maybe we need more discussion on this, but here are some obvious
ones:
o Interim meetings can be extremely productive in the right situation. Or
completely unnecessary in others. It is difficult to find a situation where the
planets are properly aligned such that several working groups are in the right
phase to have an interim meeting, and that the co-location with some event
makes sense for all of those working groups. In other words, the success of a
LIM seems somewhat unlikely, though there may be success for individual WG
interims.
o Co-location with RIPE appeared useful. I agree with you Joel that tighter
packing would have made a difference. I met some people who noted they will not
attend, but probably would have attended if it was during the week. Co-locating
individual WG interims with RIPEs and NANOGs seems like a useful concept to
consider in the future.
o LIMs will not create a new big funding source for the IETF. We should also
right-size our organization for the task at hand. 30, 50, or even 100 people
could probably be handled as part of the RIPE meeting, and might have been
something that the RIPE registration system and agenda could have accommodated,
or have someone sponsor a room and leave the rest to participants.
o As usual, hallways matter more than the official meetings. Put the right
people in the same building, and I'll attend, even if the topic would be
International Charactersets for EBCDIC-based FTP.
o Saturday is not a good day.
Detailed comments:
The LIM was the attempt that I am aware of an interim meeting
scheduled by IETF management for the purposes of accumulating interim
Does not parse. Did you mean to say that LIM was the only attempt of its kind?
A couple of years ago we also had the Malta proposal which failed (possibly due
to the downturn, or possibly due to the same issues that made the Amsterdam
event surprisingly small).
meetings in a common location rather that scheduled by working-group
participants, chairs nad coordinating ADs.
s/and/nad
It is, my understanding that discussion of the possbility of a LIM
style meeting occured early2011 if not before.
s/possbility/possibility/
s/early2011/early 2011/
There has been discussions of something like this over the years. FWIW I am not
aware of the exact details of when and how this LIM discussion started. Maybe
Ron or someone else from the current IESG could explain it.
The stated rational for targeting v6ops
involvement in a large interim was the volume of work that we process
during and between meetings.
Interestingly, the amount of work in the interim meeting itself did not seem
match this as well as I would have expected. We can speculate about the causes.
I came prepared to talk about certain topics, but the draft authors in question
were not present. Perhaps an isolated case, but had the impression that you
guys dealt with only a subset of the work that v6ops meetings in IETF deal
with. Do you agree?
Superficially only a fraction of the v6ops attendees are
represented by the interested segment however when the numbers are
mapped against active participants and draft authors, interested
participants in the interim represent a bigger purportion of that
group
Missing ".".
I think we have a participant list. It would be useful to have an understanding
of how many % of the people who presented or made mike line comments in IETF-84
were present in the interim.
One osbervation that I would make about the interim submission
observation
V6ops attendance at an IETF meeting is typically in excess of 200
attendees. An interim meeting that attracts 25 of those and
minuscule remote participation is necessarily exclusionary by default
if not deliberately. If useful work that advances drafts, gets done,
is that exclusivity a bad thing? It's not useful for measuring
meaningful consensus.
They are different meetings. And for sure the interim meetings do not help in
measuring consensus. However, at their best they are meetings where closely
knit groups of core developers can progress their work in a high-bandwidth
collaboration environment. I have yet to see an interim meeting that would
actually represent any meaningful consensus measurement point (to the extent
any physical meeting can; the list is what counts). Or perhaps the Montreal
IPv6 transition meeting was one, attended by close to a hundred persons. In any
case, this does not necesssarily render the interim meetings useless, just that
you have to use them as a different tool. Progress work. Move forward in
design. Write specs. But far less consensus measurement, far less various
experts from different domains, etc. Useful early in the work, not so useful
later when finetuning something.
Area director's were rather well represent at the LIM, While the
attendance of both of our Directors was appreciated I'm not sure that
it's a good use of their time. In particular if the frequency of
these events were fixed as some rate in the future, this represents
an additional workload for which huge benfits due not appear liekly
to ensue
There were also (at least) two IAB members, both of whom were present on site
for other reasons as well.
Jari