IETF 63 Plenary, Wednesday

(Notes taken by Mirjam Kuehne.)

1. Brian Carpenter, chair of the IETF, welcomes the IETF participants
and presents the agenda.

He shows a number of statistics:  At IETF 63 there are 1454
participants; that's 150 more than at the last mtg, but 15 less than at
the meeting one year ago.

People from 36 countries attended the meeting. Only 1/3 are from the
US. It is worth noting that many people from the ITU-T are present at
this meeting.

4 new WGs were established, 22 WGs were closed (Applause)



future meetings:  Fall 2005: Vancouver, Spring 2006: USA

2. Pascal Viginier, Executive President R&D, France Telecom

Pleased to welcome IETF in France and in Europe

Briefly describes the France Telecom network. IP is one of the 5 new
business technologies. New Experience of telecommunications (NExT) will
be a big shift from pure network to services (content, business
everywhere, service everywhere). In order to deliver that we need
reliable IP networks, seamless networks, convergence, wireless (most
difficult to implement), also need to provide voice, data, images to
customers. Need dedicated protocols to guarantee best quality services.
Need security, QoS to make things as easy as possible and seamless for
the customers. We need to manage complexity in the network and
protocols so that it is easy for the customers to use.

Building a full integrated network with best advanced protocols,
including an increasing number of machines in the next few years,
probably even more machines than people will be connected. They will be
connected with IPv6.

Standards is key. We love standards, we need them for interoperability.
The speed of standardisation is important. Encourages the IETF to
continue the way they are working. We would like to see the IETF
increase cooperation with other SDOs, especially with the IEEE, but
also with the ITU, 3GPP and other mobile SDOs.

Viginier thanks all the helpers from France Telecom and all the
volunteers and the sponsors (Cisco, Juniper, Renater).

3. Hideka Tiku shows some network statistics.  For more details, see
presentation

4. Francois Mercier explains the architecture of the network For more
details, see presentation

5. Lynn St.Amour, president and CEO of ISOC

On of the roles of  the ISOC president is to appoint the NomCom chair.
Very happy that Ralf Droms has accepted the appointment, as a new chair
of the nomcom.

Lynn thanks Danny McPherson for his work as NomCom chair last year.

Mentions the BoT meeting Friday and Saturday this week. The BoT
meetings are open. Introduces all the BoT members. Introduces Daniel
who is a new BoT members. He is also the chairman of the Jon Postel
Award Committee.

6. Daniel Karrenberg describes the Jon Postel Award and introduces the
Award Winner for 2005: Jun Murai.

Jun Murai thanks ISOC for the award and mentions that he is very
pleased to see that there are so many attendees from JP and from Asia
at the IETF nowadays. Remembers the time when he was the only attendee
from Asia at the IETF. Also remembers the time when he installed the
first root server that was located outside the US.

7. Lucy Lynch, chair of the IAOC:

Introduces the IAOC members Describes IAOC responsibilities and the
structure of the IASA, IAD responsibilities

Recent activities:
 -       BCP published in April
 -       Official IAOC seated in April
 -       IAD hired end of May
 -       In June official web site went live:  koi.uoregon.edu/~iaoc
 -       Nomcom appointees announced in June
 -       Ongoing work:
           o Working with CNRI on transfer of ietf related IPR "the trust"
           o Potential service contract with NeuStar

describes the trust issue.  describes the potential service contract
with NeuStar

If questions, send mail to iaoc@ietf.org, iad@ietf.org


8. Ray Pelletier, IAD

Excited to take this position, because the work of the IETF is very
meaningful and he is happy to be part of it.

The job is to improve throughput and to reduce friction and add
transparency.  BCP 101 shows that we cannot continue as usual. We have
to have change.

Asks people to send suggestions on process issues to his e-mail address
regarding meetings, WGs etc.


9. Bob Braden: RFC editor report

Trends of submissions per month is up; There is still a big backlog,
but hiring more staff to cope with the growth.

Constantly working on improving performance in various ways (see
slides).

10. Barbara Roseman, IANA Ops Manager: IANA Status update

Target to process I-Ds in 30 days is not quite reached yet, but getting
there.

Port requests backlog has been addressed, a new person was hired and
performance went up significantly.

Ray Pelletier would like to thank Foretec staff for their help in
putting this IETF together and all the sponsors again.


10. Henrik Levkowetz: The IETF Tools Team

All the tools team materials: tools.ietf.org (presentation is also
there)

There are 11 currently work items.

item 6:  Pointer to ww1.tools.ietf.org/wg for better draft management.

item 7:  Agenda will be html-ized


11. Brian Carpenter: Next Steps (see his slides)

IETF attendees and core attendance went down significantly over the
last 4 years, but not as much as the NASDAQ

Achievements
 -       IETF mission statement
 -       Tools team
 -       Edu team
 -       Improvements in IESG doc processing

In progress:
 -       a number of efforts to improve transparency
           o retreat and other minutes

IESG is taking transparency very seriously

IAB progress:
  -       governing RFCs regarding liaisons with other
          orgs are published

still need clarity on how "requirement" liaisons from other
organisations are fed into the IETF process.


In progress in the IETF
 -       IPR WG basically done
 -       NEWTRK (disagreement between WG and IESG on ISD)
 -       IANA considerations discussion
 -       What are the IETF requirements for spec publications. 
         There will be a techspec BoF in Vancouver.

What's open?
 -       quality and timeliness of WG output; more to do
         on engineering practices in WG, tools can help to manage docs.
 -       Cross area review needs to be improved
 -       IESG throughput rate and role (will re-organisation make
         things just different or better?)
 -       Process documents are a patchwork and extremely difficult to
         understand
 -       IETF chair has conflicting roles, needs to be discussed


There are some very open issues related to the problems statement RFCs

World has changed, IETF process change is apparently needed. However, the
plane is flying, so we have to be very careful in changing the engine.
And what should be the forum for process change?


12. 15 minutes open microphone directed towards the IESG

James Kempf: feels that there are problems with the formats of RFCs,
they are not hyper-texted and don't use modern publication
technologies.

Brian Carpenter: suggests to write a draft so that it can be
discussed.

Spencer Dawkins: appreciates Brian's summary, including the personal
summary. Thanks IESG to put together the DISCUSS document. We need to
get really good at early reviewing docs before last call. Concerns that
when talking about the 2 or 3 step standardisation process conflicts
with the 0 step process we appear to use.

Bill Fenner clarifies that the 30% number (of drafts that get no
DISCUSSes) is slightly wrong due to a programming error.

Discussion about the DISCUSS topics.

John Loughney: thinks that the DISCUSS has improved the documents.
Would like to see earlier reviews, more cross area review etc.

Sam Hartman: wonderful goal, but a lot of work to get there. Where can
you contribute to earlier reviews: guidelines on handling earlier
reviews; there are a lot of tools needed (needs to find out which docs
are in LC in WG); getting people to do reviews (WGs have to take care
of that and make sure they have enough people to reviews).

Some comment that the IESG needs to function better

Margaret Wasserman asks everyone who thinks she (or any of the IESG
members) has done something wrong to actually come to her directly and
talk to her about it. This happens extremely rarely.


13. OPEN microphone "town hall meeting" session.

The following issues have been raised:
 -       Project mgmt/delay
         reduction for drafts (Henning Schulzrinne)
 -       Process change ideas
        (Spencer Dawkins)
 -       Does the community believe process change is
         needed (if so, at what priority)? (Bert Wijnen)
 -       What is the best way to change the process? (Brian Carpenter)
	o At least how do we focus the list discussions?

(see Brian's slide)

issue 1:  Henning Schulzrinne: perception to spend a lot of effort to
reduce the delay in draft publications, we have not been successful.
Don't even have the tools to find out if we have been successful.
Thinks it requires a change to have a notion to take a passive
conference approach. The authors is the one who needs to push. Moving
to a more project mgmt approach. Be more customer driven. WGs needs to
develop processes and rules.

Leslie Daigle: that means to move to a more planned and managed process
with better traction of activities.

David Black: a running code observation. As a WG chair when I am paying
attention to drafts then they tend to move faster. WG chairs need to be
better process managers.

Keith Moore: strikes him that many people come from a product producing
environment. This model cannot be applied to the IETF, because we are
producing long-term standards, need to be good for many many years. We
need to do better engineering.

Issue 2 Spencer Dawkins: John Klensin had two drafts submitted and
proposed a standards review panel There is a lot what the communities
can do to improve the review process.

Ted Hardie: good first step would be to do a functional
differentiation. Suggests that the IESG is not the body that determines
the outcomes. Create a process change process that would not have the
IESG as the review body that review process change.

Margaret Wasserman: 1. comments on review panel draft. Believes that
dividing the management duties and the doc review would be a good idea,
but this is not what the drafts say. The draft still keeps all the
responsibilities within the IESG. Disagrees with the previous speaker:
should not let the WG decide about process changes. 3. when I joined
the IESG did not stop to be a member of the IESG, we should stop making
that distinction.

Discussion what would help: rules, tools,

Sam Hartman: I think we got too much process stuff going on and get the
process stuff and the technical work going on. There are a lot of
people who are involved in the process if they like it or not and for
them the processes and the process changes need to be workable. Do we
need to do all these process proposals at once? Couldn't we serialise
this?

Discussion about responsibilities of WG chairs vs. doc authors.

Bert Wijnen: has been on the IESG for 8 years, certainly in the last 5
years many process changes were happening. Not against change. The
review board may be fine, because it may unload work of me. Expects
that we will continue with change discussions over the next few years.
It's a question of priorities: should the ADs work of the backlog or
spend time on change processes?

Charlie Perkins: process change is needed, maybe not every AD has to be
involved in this discussion.

Harald Alvestrand: process change is needed. Persons who execute the
process are not able to see the process from the outside. They are not
the persons who need to decide. See admin restructuring: figure out who
has the admin responsibility and let them run with it and accept the
result. Better way than asking the IESG to review more BCPs.
(Applause).

?: process change is needed; biggest discrepancy between IESG and
community is that IESG is too overloaded, they don't have the time to
be an active part of the community. Needs to change the communication
between the IESG an the WGs. Potential way to do that to have the WG
chairs get together and give a presentation to the AD rather than
having them to read the documents from scratch.

Brian Carpenter: will dig deeper into the statistics to understand the
delays.

Leslie Daigle: entire community needs to find a way to agree on how to
pick what process changes we need to work on them in what order.

Suggestion to work on process changes in the plenary.

Pete Resnick: suggests to find a way so that the IESG can stop
reviewing docs and be an AD (help the WGs to get their work done)

John Klensin: if our method of process change is too slow we never get
anything changed. Drafts were written because we were trapped between
two problems: if you write a draft that describes high level concept,
people will say the details are missing; if you write a draft
describing all the details, people will say the higher level concept is
missing. If we can't get off of that trap, we won't go anywhere