IETF Operations and Administration Plenary Minutes

   (Edited from notes by Mirjam Kuehne)


The plenary session was divided into of two parts.  The first part
covered the normal administrative and operations plenary topics.  The
second part was dedicated to the routing and addressing problem status.


Part 1: IAOC/IESG Plenary

Agenda:
   1. Welcome
   2. NOC Report
   3. Host presentation (Neustar)
   4. IETF Chair and IAD Reports
   5. NomCom Chair (Andrew Lange)
   6. Introduce New IESG and IAOC Members
   7. IAOC Q&A 8. IESG Q&A

Welcome (Brian Carpenter)

NOC report (Morgan Sackett, VeriLAN) (see slides)

Host presentation (Jon Lindberg, VP of Secretariat Services, NeuStar)
(see slides)

Brian Carpenter presented Jon Lindberg with a plaque of contribution to
NeuStar for hosting IETF 68.  This gesture was recently introduced to
recognize the contribution made to the IETF by the meeting host.

IETF Chair and IAD Reports (see slides)

NomCom Chair (Andrew Lange) (see slides)

Brian welcomed all the new IAB, IAOC, IESG members.  Brian passed the
dots from his badge to the incoming IETF Chair, Russ Housley.  Russ gave
a presentation (see slides).

IAOC Open Microphone

Lucy Lynch, the outgoing IAOC chair, recognized everyone who is leaving
the IAOC and welcomed the new members.  Then, Lucy announced that Steve
Crocker, the author of RFC 1 and many others, and his brother Dave
Crocker, an author of many RFCs as well, are the first two RFC authors to
sign a non-exclusive license to the IETF Trust granting the IETF Trust
copyright interests in the RFCs in which each was an author from the date
of creation.  The IETF Trust can sublicense, reproduce, publish, modify,
translate, excerpt, combine, and create derivative works of the RFCs.
Steve and Dave stepped up to the podium and encouraged others to do the
same.

Someone asked if they could do it at the meeting, and Lucy said that the
IAOC office hours could be used to do so.

Fred Baker: would it be rational to send a digitally signed email to Ray
Pelletier in lieu of a signed letter?

Lucy: would have to take this up with the IETF Trust lawyers.

Eric Rescorla complained about the cookie-free break in the afternoon.
He complained that the cost for cookies at all breaks was unknown, so a
community discussion was not possible.

Ray: will ask the community as part of the survey after the meeting
including the rough costs for additional food supplies.

John Klensin: policy decisions should be made by the community and then
the IAOC should implement them.

Lucy stated that she believes that the decision about cookies is
micromanagement and not a policy issue.

John: as we look at the portion of the meeting fees that does not
directly cover the meeting, how does the community provide input on how
those funds will be used?

Lucy: good question and will be put on the agenda for Chicago.

Sam Weiler: why aren't we asking for letters from RFC authors for
documents with RFC numbers greater than 2026?

Brian: agrees with Sam that it should not stop there. But it seemed to be
more urgent to focus on those where the rights are unclear.

Phil Hallam-Baker (referring to the cookie discussion): the larger
problem is that you might think that what you are buying is the cookie,
but you are actually buying the meeting room. All the little incidentals
are important to the hotel. Phil doesn't think we want community input on
the details of the contracts with hotels.

David Kessens: would like to ask for more microphone control. The time is
way too valuable to discuss micromanagement issues like cookies. David
does not want this issue to come up again during the plenary sessions.
This can be done on the mailing lists.

Daniel Karrenberg: thanked the IAOC for an excellent and professional job
over the last three years.

Leslie Daigle: agrees and mentions that this conversation would have been
inconceivable a few years ago. Congratulations to us all!

IESG Open Microphone

Spencer Dawkins: thanked Andrew for the NomCom report. The incumbent
syndrome was already mentioned in John Klensin's report last year.

Russ: this topic already on the agenda for the IESG retreat.

???: was quite overwhelmed by all the documents that need to be read
before the meeting. We seem to get to a write-only mode: a lot gets
written, but there is not enough time to read all of it.

Russ: a balance is needed.  When the queue was closed too soon, people
would publish their documents on their own web sites. He encourages
people to write more short and succinct documents.  Of course that take
more time.  Mark Twain said: "I didn't have time to write a short letter,
so I wrote a long one instead."

Pete Resnick (referring to the comment Russ made above on the NomCom
issues being on the IESG retreat agenda): is it correct that the IESG
will look into NomCom issues?

Russ: The IESG is going to discuss the recommendations made by the last
few NomCom Chairs.  If there is some consensus about the best way to make
improvements, then the suggestions will be shared with the IAB and the
rest of the community.  Obviously, the IESG cannot make any changes or
decisions on its own.


Part 2: ROuting and Addressing Problem (ROAP) Status
        (Internet and Routing ADs on stage)

Brian gives an overview of the ROAP (ROuting and Addressing Problem)
(see slides)

Open Microphone

Ted Hardie: thanks Citizen Carpenter for the presentation, but has the
feeling it was way too calm. Ted would like to ask the IETF to be bolder
in tackling this problem. Ted doesn't see evidence that this problem is
taken seriously enough. There are people who think that separating the
locator from the identifier is an easy solution. That is not the case. A
lot of things will break. Don't assume that any of the current
suggestions are the solutions.

Ross Callon: there are different types of panic. We are not going to come
up with a solution without costs. We don't understand yet what the
tradeoffs are. It is good to know that there are engineering solutions to
keep the Internet running in the meantime, but we need to start working
on some fundamental changes. He has been talking to a lot of people since
the IAB Routing and Addressing Workshop in Amsterdam. Ross agrees that
this needs to be taken seriously.

Spencer: What about BGP wedgies and dynamics (described by Tim Griffin
and Geoff Huston in RFC4264)?

Ross: regarding BGP dynamics, there are some things that can be done, for
example with operational practices, what policies one allows, tweaks in
implementations or in protocols. There are discussions about this right
now. There is not yet a lot of public discussion, but it will start
tomorrow at the open area meeting.

Spencer: on the identifier/locator split, please be sure that you get
enough input from the applications people.

Mark Townsley: point very well taken. That was one reason why we
organized this as part of the plenary. We want to make sure that all
interested parties are aware of the topic being discussed.

Dave Ward: Dave Meyer was the catalyst of bringing this problem to light!
(applause)

Dave Meyer: a lot of colleagues helped. We are 15 years into this. Are we
going to go another 15 years? Russ was talking about incremental
improvement. Are we not doing this already for BGP? This is a
controversial topic. It will be hard to get IETF consensus. This will
also involve process. Do we have any way forward in our process to deal
with this?

Ross: Dave helped get us moving. We have now started moving, though not
quite in the same direction yet.

Dave Ward: emphasizes that the end game is not improvements to BGP (yes,
this has to be done anyway). A ROAP directorate has been put together.
For now we have to fit in the processes that exist. Another aspect of the
way forward is the rechartering of Routing Research Group.

Mark: believes a lot has been done in the last 6 months: - a directorate
formed - two WG and area discussions - a plenary discussion please tell
us what more we can do.

Daniel Karrenberg: let's learn from past experiences. There was a panic
about routing table growth that gave us CIDR. We needed a quick fix, and
we did a quick fix. There was a panic about IP address space running out.
That gave us IPv6, and some of the 'features' IPv6 brings, is because we
rushed it. Daniel encourages us to look at past experiences and learn
from them.

Timothy ???: something is happening that goes in the right direction. Not
clear if we figured out what the right processes are to make
architectural changes to the Internet. Please think carefully how this
can be done. It is too important to leave this to the IESG or to the IAB.
One of the key issues: almost any architectural change will have tricky
costs, but especially implications in security. We need to include
security implications early on in the process.

Jim Bound: points to the IETF Journal. It includes a pointer to a the
'new-arch' document. He encourages everyone to read it. We need to bound
the solution space. There are solution-spaces that work in 2 years, in 5
years, in 10 years. That boundary has to be understand upfront.

Phil Hallam-Baker: maybe we don't need to develop new architectural
principles, but old architectural principles that are not currently
applied. Phil mentions a few examples.

Tony Hain: talking about solutions over different time frames. But we
also need different solutions that work in the same time frame. There is
not one solution or tool that will solve all problems. Different sets of
tools maybe solving different parts of the problem. Trying to solve
everything with BGP may not be the right direction.

Ted Seely: Brian made a comment that the business decisions are not part
of the IETF discussions. Disagree. This system is very fragile from a
business perspective.

???: extremely concerned, not only from an economic perspective, but also
from a process perspective. Look at how long IPv6 takes to get deployed.
It will take us several years to develop, implement and deploy new
architecture. FIB size is the fundamental issue.

Ruediger Volk: looks like we are neglecting what Ted Hardie said: he
agrees with the slides and agrees that panic would be bad, but we need to
seriously concentrate on this problem and continue to work on it. The
operators and users will not be able to pay continuing updates of
hardware. We have to make sure we deal with the architecture in some way
that does not lead into high costs and short lifetime of equipment.

Alain Durand: success to good governance is to share the pain. Make sure
that people will benefit from the solutions will be the ones who pay for
it. We also seem to have already constricted ourselves to a small problem
space: routing (and small incremental changes) and addressing (big
fundamental changes).

Sue Hares: has been working in routing for over 20 years. To make
efficient forward progress, we need to do a lot of data mining first, but
we need to do it quickly.

Margaret Wasserman: for short term solutions, she understands that we
need to concentrate on routing. But for the longer term solutions, a lot
more will be involved, probably all areas. Understands that the ROAP
people are trying to do that, but would like to see some more evidence of
it (e.g. involve them in the ROAP directorate). We don't really have a
common understanding what the IP architecture. Not sure if we are not
actually moving the problem. Believes we need to understand the
implications of the identifier and locator split.

Dave Crocker: good, animated discussions. He believes that people have
not heard Dave Meyer correctly. The list of problems on Brian's slides
was way too long. Suggestion: find a smaller set that the Internet Area
can solve, find a smaller set of problem that the Routing Area can solve.
It can be that they will then come up with solutions that will also solve
other problems in other areas.

Bob Hinden: guess that if we're trying to make fundamental changes and
keep all the existing business models. The Internet got started by
breaking the then existing business models completely. It could be that
this will also happen today. We should not be constrained by the way
people do business today.