By: Mirjam Kühne
Date: May 7, 2006
IETF chair opening
Brian Carpenter, the chair of IETF, welcomed everyone to the plenary. He presented attendees and document statistics (see Fact & Figures on page 2). This time some people had problems receiving their visa in time. Thanks to ISOC’s urgent intervention with the US State Department, this could be solved in most cases. It is still damaging to the IETF’s openness as Brian pointed out.
NOC and Host Report
Rich Osman, in charge for the NOC at the IETF this time, showed some network statistics. He thanked the many volunteers who helped during the week and did an especially good job, when the NOC had to be moved within 20 minutes on Sunday due to rain in the NOC.
Paul Murdock, representing Nokia, the host for IETF 65, gave a presentation showing some of Nokia’s latest multimedia developments.
The IASA Report consisted of four presentations:
– RFC Editor report (Bob Braden)
– IANA report (David Conrad)
– IASA operations report by the IAD (Ray Pelletier)
– IAOC general report: IAOC Chair (Lucy Lynch)
Bob Braden reported that the RFC Editor has made significant progress towards reducing the publications backlog reported at the last few IETF meetings. At current rate the backlog will be completely gone before June 2006.
David Conrad, the General Manager of IANA, reported that at the last meeting a number of issues were identified, mostly related to the lack of consistent management, understaffing and too complicated processes. Many of these issues have been addressed which is visible in the improved IANA performance.
Before the next meeting, IANA is planning on clearing up the current backlog and hiring two additional staff, one of them fully dedicated to IETF liaison. IANA will also work on improved tools and will publish IETF-related processes and data on IANA performance.
David thanked Michelle Cotton, who has now been working for 6 years at IANA and was at times the only IANA person.
Ray Pelletier, the IETF Administrative Director, showed an overview of the IETF expenses for the Fiscal Year 2005 and the budget for 2006. He then gave a status report of the secretariat services after being in place for 90 days. The transition from Foretec to NeuStar Secretariat Services went well, new tools and infrastructure have been set up as well as longer-term planning for future meetings.
Lucy Lynch, the chair of the IETF Administrative Oversight Committee, reminded the plenary that after IETF 64, the IAOC had issued a Call for Consensus on the formation of an IETF Trust. The founding parties included CNRI, ISOC and the IETF. Some substantive issues were raised and the documents were modified to address community concerns. In December consensus was reached. The final IETF Trust Agreement can be found here(pdf).
Consequently a number of actions were taken to round up the Trust Agreement. In recent months, the IAOC has been focused on
- the development of an RFP for the RFC Editor services
- conducting the first annual review of the IAD
- future IETF meeting planning
- monitoring the US Department of Commerce (DoC) RFI regarding IANA services
Lucy clarified that the RFC-Editor RFP will be an open RFP and that there will be a request for interest first. Leslie Daigle added that she sent the planned process to the IETF mailing list recently.
Major projects for IASA in 2006 entail:
- IAD annual review and goal setting in June 2006 (IAOC)
- development and Execution of the RFC-Editor RFP (IAOC)
- managing IANA issues: DoC, contracting etc. (IAD)
- community relations per BCP101: minutes, financial updates, active listening (IASA)
- formalising the budget process (IASA)
- IPR related issues (Trust)
- managing archives and assets (Trust)
Updating the Tao of the IETF
Susan Harris gave a short summary of the main changes made to the Tao document (see more details on page 11) and asked the community, especially IETF Newcomers for feedback.
Ralph Droms, chair of the NomCom, listed the nominees for the open IESG, IAB and IAOC slots. It was a difficult task this time, because of the creation of the new Real-time Applications and Infrastructure (RAI) area and a resignation on the IAB.
The NomCom schedule, as laid out in RFC3777, turned out to be tight. Ralph and the NomCom proposed to review this schedule significantly. The tools developed together with Henrik Levkowetz and the tools team were very helpful and the development of additional tools for I* nominations and NomCom volunteers are being considered. Henrik added that the time schedule problem might also be solvable with the help of new tools.
Jari Arkko : Internet Area
Dan Romascanu: Ops & Mgmt Area
Ross Callon: Routing Area
Sam Hartman *: Security
Magnus Westerlund: Transport Area (2 yr)
Lars Eggert: Transport (1 yr)
Cullen Jennings: RAI Area
Jon Peterson **: RAI Area
Incoming IAB Members
Leslie Daigle *
Eric Rescorla *
Incoming IAOC Members
Ed Juskevicius *
* returning incumbent
** moved from Tarnsport
“Thank You” to all departing IESG and IAB members
Fred Baker, the chair of the ISOC Board of Trustees (and former IETF chair) thanked all departing IESG members for their time on the IESG. They have done a great service to the IETF. Brian Carpenter echoed these thanks.
Leslie Daigle, chair of the IAB, thanked all departing IAB members in the name of the entire IAB.
Lynn St.Amour, CEO and President of ISOC, handed out plaques to Jonathan Rosenberg, Patrik Fältström, Pekka Nikkander, Pete Resnick and Bob Hinden as thanks for their service on the IAB and to Bert Wijnen, Allison Mankin, Alex Zinin, Scott Hollenbeck, Margaret Wasserman for their service on the IESG.
Alex Zinin, one of the departing IESG member, reflected on the IETF in the last 4 years. He felt that the IETF has become a more open and cooperative organisation that is more innovative and more international. It is also more operations-aware than a while ago. At the same time he felt the IETF has also become more formal and process oriented.
Alex pointed out a number of things he learned during his time on the IESG and gave the following recommendations to new IESG members:
- trust your first impressions
- you are a newbie – use that! (you bring the background from the community)
- don’t allow politics to rule avoid bureaucracy burn-out
- if it stalls – take it to the public (community can help!)
- remember who your real boss is (the community!)
- remember why you did it (keep your nom-com questionnaire)
- know when to stop
Bert Wijnen, another departing IESG member, who served on the IESG for 8 years added that even though we are all here as individuals, he would like to thank his employer, Lucent Technologies, who basically sponsored his work for many of those years.
Bert also thanked the community for the help and support. He mentioned that he got a little worried lately about the IETF and the IESG (referring to issues like blocking people from mailing lists, a lot of the process issues etc.). He would like to see the IETF continue to focus on technology. As a thank-you to the community he sang two Dutch songs which was enthusiastically welcomed by the audience.
Nominet (present at IETF 1): “One of the things about the Internet is that it was designed to be used by experts. We have gone from a benign environment where we knew everyone on the Internet to an environment where we have cyber-stalkers and cyber-terrorists. It is a different world. The web has changed things. I think for the good, because lots of people can get their voice out, but part of it is that we are diluting the truth… The Internet is the best system for spreading memes that we have ever seen.”
|Open mic sessionBrian Carpenter welcomed a suggestion to present IESG statistics to the community similar to the IANA and RFC queue statistics. He mentioned that the IESG had an ‘Efficiency Retreat’ and set some targets for itself. They should certainly be presented. Bill Fenner and Allison Mankin are working on tools to present them.
This was followed by a discussion about the perceived lack of food – notably cookies – during the IETF 65 meeting breaks which some community members took as a sign that the IETF was in very good shape if that much time is spent on such a topic during an IETF plenary session. Ray Pelletier admitted that the numbers were underestimated and were adjusted for the rest of the week.
Another topic that got raised were individual submissions to the RFC Editor. One author felt that the technical review board does not have sufficient expertise in one particular area. Joyce Reynolds, representing the RFC Editor on the IESG, noted that the RFC Editor is always trying to make the best effort to find the best reviewer on the review board, but that they cannot guarantee to have expertise in all possible fields. All reviewers are volunteers.
Complexity, a topic raised at previous IETF plenary sessions, was brought up again. Some people were worried that an increasing amount of complexity is exposed to the end-user, especially in the area of security. This can lead to users not deploying security standards. Sam Hartman, one of the security ADs, agreed and said “We don’t only need to make sure security mechanisms are complete, we also need to ensure they are deployable.” He suggested to develop more show cases and examples in order to demonstrate how security mechanisms can and should be used.
At the end of the plenary, the role of the IETF chair was raised. There was some concern that the IETF chair has too many roles to fill:
- IETF chair
- IESG chair
- General Area Director
One speaker felt that chairing the IESG might be the most important role for the IETF chair. Other tasks of the IETF chair could possibly done by others.
There was no agreement on this topic and in closing Brian pointed out the importance of the external role of the IETF chair.
Welcome and Introduction by the IAB chair
Leslie Daigle opened the technical part of the IETF plenary and announced that she was honored to be re-appointed as IAB chair for the next year.
Aaron Falk, chair of the Internet Research Task Force (IRTF), described some recent IRTF highlights. A number of Research Groups (RGs) met during IETF 65. One of them was the Anti-Spam RG which also had a review with the IAB. The two new research groups (the Transport Modelling RG and the Internet Congestion Control RG) are making good progress. There is a charter under development for a MANET RG and for a group on scalable small-group multicast.
The document “IAB Thoughts on the role of the Internet Research Task Force” is about to be published as an RFC. In addition to that the IRTF published draft-irtf-rfcs-00.txt to describe a process for IRTF RGs to publish RFCs.
|IRTF Technical Presentation – End-to-End Research Group
Karen Sollins, one of the co-chairs of the IRTF End-to-End (e2e) RG presented the status of that group. The e2e RG has existed since 1984 and is limited to 20 members with rotating, closed membership. The focus of the RG is on End-to-End services and protocols. There are 1.5 day long meetings 2 – 3 times a year. The group is set up as a forum to exchange ideas. Topics which have been discussed in the past include:
At their last meeting, the e2e RG posed the following question: ‘How might the computing and communications world be materially different in 10 – 15 years, and how might we define a research agenda that would get us to that world?’
Answering that question requires a vision of the future Internet. The e2e RG concentrated on the conceptual vision rather than on possible technical approaches to get to that future Internet.
professor at UCLA (present at IETF 1): “We need to develop a vision. When the network was still small, we did not really need a clear vision… When we were slightly off track, we said ‘ooops, let’s adjust it.’ But when the system is huge, it is much harder to adjust. If it is not steered into the right direction at all times, it will be very hard to turn. Therefore, I believe that today, it is essential to have a clear vision, because the network is this gigantic thing that is moving along and that we have to steer.”
Karen encouraged everyone to develop visions as well as architectural requirements that would bring us there.
More details on this work can be found in the following publication: ‘Making the World (of Communications) a different Place’ by Clark, Partridge, Braden, Davie, Floyd, Jacobson, Katabi, Minshall, Ramakrishnan, Roscoe, Stoica, Wroclawski, Zhang in ACM SIGCOMM CCR 35 (2), July 2005, pp. 91 – 96. A copy of that document can be found here(pdf).
IAB Technical Presentation – Distributed Hash Tables
The IAB has regular informal, internal tech-chats. On a recent tech-chat about Distributed Hash Tables (DHTs), it became apparent to the IAB that there was material that would be interesting to share with the community – hence this plenary technical presentation.
Eric Rescorla, an IAB member, described the overall concept of DHTs, explained the difference between search and lookup and gave some examples . He further described different security mechanisms, their applications and the open issues for each of them.
In closing Eric noted that the work on Distributed Hash Tables is mostly research today and not really relevant to normal users at the moment even as they are being actively proposed for use within various protocol standards.
Not related to the above topic, but in good IETF tradition to combine serious work with some fun, various IETF participants stood up and gave Eric bags of cookies (referring to him complaining at the Administrative plenary about the lack of cookies during the breaks.
Leslie Daigle gave an update of the IAB activities of the year.
The IAB identified 3 areas of interest last year:
- Architectural Perspective
- Unwanted Traffic
- And, of course there is always a “misc” category !
Regarding IPv6, the IAB started a dialogue with the operational community. A special thanks to Dave Meyer, a member of the IAB, who opened channels to the operations community by organising BoF sessions on IPv6 Multihoming at operators meetings (NANOG, APRICOT, RIPE). This effort will be continued. In addition to that the IAB has set up an IPv6 ad hoc committee.
On the topic of Architectural Perspective, the IAB has published a number of documents:
- RFC 4101: Writing Protocol Models (June 2005)
- RFC 4417: Report of the 2004 IAB Messaging Workshop (Feb. 2006)
- RFC 4367: What’s in a Name: False Assumptions about DNS Names (Feb. 2006)
The architecture-discuss mailing list : [email protected] was also established for broader community discussions on architectural issues.
The IAB IDN ad hoc committee will be shutting down.
At the beginning of March 2006, the IAB hosted a workshop on the topic of ‘Unwanted Traffic’, with people from various backgrounds: backbone operators, enterprise managers, researchers, software developers.
The goal was to get cross-area communication and to raise awareness. As a main result from the workshop it has been recognised that the IETF needs to continue to build mechanisms to address the usual security and unwanted traffic issues while at the same time mechanisms have to be developed to respond to more serious, more organised security threats. The solutions might not be the same. A full workshop report will be published as an RFC.
At the end, Leslie listed some other ongoing and administrative IAB activities:
- Tech Comms ad hoc committee (to work with ISOC on a series of documents)
- IETF-related process and liaison documents
- RFC 4089: IAB and IESG recommendation for IETF Administrative Restructuring (May 2005)
- RFC 4052/BCP 0102: IAB Process for Management of IETF Liaison Relationships (April 2005)
In RFC queue:
- Jefsey Morfin, suspension from ietf-languages mailing list
- IAB annulled IESG decision
- Julian Mehnle, MADRID documents
- IAB upheld IESG decision
Masters Public Policy and Computer Science at Univ. of Maryland (Newcomer at the IETF): “One of the things that attracted me to come here – which is probably a double-edge sword – is the fact that it is so focused on the working groups. That makes it interesting, because you’re actually doing work. On the other hand it is also what makes it inaccessible for Newcomers, especially if you haven’t been reading the mailing lists for a while.”
The discussion around the RFC Editor has been continued during the IETF 65 GENAREA meeting. The overall timeline was sent to the IETF mailing list, as well as:
– IAB documents.
– current IAB Internet-Drafts.
– IAB Minutes.
As all the IAB members came up on stage to host this session, a video of “virtual” Vint Cerf sharing congratulations to the IETF’s 20th anniversary of the IETF was presented.
During the open mic session the topic issue was the End-to-End principle (in reference to Karen Sollins presentation on the e2e RG; see page 8).
It was noted that e2e and network neutrality today might be more a business rather than a technology decision.
While some people felt that the IETF should continue to concentrate on the development of protocols and not make recommendations about ‘good’ or ‘bad’ technologies , others believed the IETF needs to be promoting technologies and components that do support the e2e model and that it has a responsibility at that (similar to the IETF’s conscious decision not to design protocols that make wiretapping substantially easier).
As way forward, comments in the room suggested that the IETF needs to accommodate both: the e2e principle as well as business considerations.
A role for the IAB in this could be to ensure the protocols work in networks today (including firewalls).
In the ensuing discussion, people wondered how the ends were actually identified. In IETF discussions, an end is typically identified by an IP address. Not everyone agreed with that. Patrik Fältström, one of the departing IAB members said: “End-to-End should be defined as the humans and their applications (e.g. chat, rss feeds). As long as the IETF designs protocols the way it has done so far, users will always be do able to use their applications.”
Jeff Schiller suggested that a new definition of Internet services is needed. i.e. end-user security (assuming the network can not be trusted) vs. security in the infrastructure (providing reliability and availability for the end-user).
The web service community is one example of this distinction: because they could not trust the network, they developed security on the xml level, as Brian Carpenter pointed out. In the network neutrality context, he recommended to look at RFC 4084 (Terminology for Describing Internet Connectivity).
All presentations given during the IETF 65 plenary session can be found here.