Internet Architecture Board


IAB Minutes 2011-01-05

Home»Documents»Minutes»Minutes 2011»IAB Minutes 2011-01-05

IAB Teleconference


1. Roll-call, agenda-bash, administrivia, approval of minutes

1.1. Attendance

Bernard Aboba
Ron Bonica (IESG liaison to the IAB)
Ross Callon
Spencer Dawkins
Aaron Falk (IRTF Chair)
Russ Housley (IETF Chair)
John Klensin
Olaf Kolkman (IAB Chair)
Glenn Kowack (RFC Editor Liaison)
Danny McPherson
Jon Peterson
Andrei Robachevsky
Lynn St.Amour (ISOC Liaison)
Dow Street (IAB Executive Director)
Dave Thaler
Hannes Tschofenig
Marcelo Bagnulo
Hassan Zaheer (IAB Scribe)
Vittal Krishnamurthy (IAB Scribe)

1.2. Agenda

This call was primarily reserved for a discussion of the RFC Editor model and next steps in the RSE definition. Two other agenda items were added due to recent events: ITU Q5/SG11 Liaison and the Internet of Things (IOT) workshop.

1.3. Administrivia

No administrative items were discussed.

1.4. Meeting Minutes

No meeting minutes were reviewed during this call.

2. ITU Q5/SG11 Liaison Discussion

The IAB has been working with the IESG to draft text for a liaison statement to the ITU regarding work on the Q.flowstatesig Recommendation. There is a new Internet Draft that summarizes the flow state signaling protocol, and its encapsulation in a GRE tunnel with a new Ethertype: draft-roberts-fsasignalling-01.txt

The issue being considered is whether the Q.flowstatesig proposal maintains a clear enough separation between its new protocols and existing TCP/IP messages so that both will operate properly. The assignment of a different Ethertype is an important component here, but the current draft also uses the term TCP to describe a modified transport layer that differs in subtle ways from existing TCP. The resulting change in the service model provided to higher layers is likely to cause problems for applications if this ‘new TCP’ is not clearly differentiated.

There is both IAB and IESG support for progressing this work within the IETF, although not yet consensus on the specific approach (e.g. , BOF / new WG vs. existing WG vs. AD-sponsored, exact contents of a charter, etc.) The IAB spent considerable time on the call discussing the options, although for several of them the IESG would have to make the final decision. Similar work has been presented to the IETF in the past, but the approach has changed sufficiently that IETF community exposure to the current work is likely small.

The board will continue to work on the liaison text, with the plan to deliver the liaison later this week.

Editor’s Note: The liaison message is now available here:

3. RFC Editor / RSE Discussion

Olaf summarized the current state, adding that the goal for today’s call was to make sure there is a shared understanding among IAB members: what the model represents, where there is support for the recommendations, and where greater clarification is needed.

Bernard raised a concern about the structure of the RFC Editor Oversight Committee (REOC), noting that some REOC responsibilities lie with the IAOC today, while others are more the domain of the IAB or RSAG. He argued that since the IAB is the appeals body, if the REOC is subordinate to the IAB then the REOC cannot be involved in contracts – a role which is implied by the statement ‘when required to participate with IASA…’ Glenn clarified that the intent was to maintain a distinction between line management and contract responsibility; i.e., the REOC would help with drafting an SOW, but the IAOC would always have control over money. The board discussed this issue, after which there appeared to be consensus. Bernard and Glenn will work together to clarify the text.

There was also some agreement that the REOC lends itself to being structured as an IAB program. Given this, Dave Thaler suggested that the draft -00 might say too much about the REOC, and that it would be better to leave some aspects of that body a little more open, allowing it to evolve over time. Andrei noted that the REOC is more than just an oversight body; it also provides community leadership. Glenn agreed, and will clarify the text.

Hannes raised the question of the ‘management level’ of the RSE. Glenn explained that the intent was for the RSE to make decisions when the policies are not clear, not to individually manage editors. For the other streams, the role of the RSE is to ensure that as the series evolves, or as issues arise from the community, the authority over the series is maintained by the community. The RSE watches the series overall, looks for broad issues, and raises them to the community. Hannes then asked about the expected seniority and experience level of this kind of manager. Glenn thought that this would be an attractive role for publication managers – perhaps a 30-something journeyman with publication experience, or someone in their 50′s who has had major career accomplishments and now wants to work on something they believe in. He noted that the prestige of the RFC Series should help attract good candidates. Glenn then noted that this is an area where the new doc differs from the previous version, in that RFC 5620 provided only a left-over position for the RSE. Now the RSE will have a better mix of authority to do the job.

After further discussion there was a sense that a shared vision was beginning to emerge. Jon Peterson questioned the move to a model with a stronger executive role, a concern also raised by Andrew Sullivan on the mailing list. Glenn will attempt to address Jon’s concern in his email response to Andrew. John Klensin raised a point about editorial workload and language clarity in the series, as more document authors will have English as a second language going forward. Russ saw the term ‘Executive Level Management’ as potentially problematic, and was concerned that the recommendation implies a senior person from the IETF community could not hold the role. Glenn responded that he thought it would be harder for someone with no publication experience to learn on the job (with no one to learn from) than for a publisher to learn the culture of the IETF (since there would be many involved individuals from which to learn this aspect). Other board members expressed general support for the direction of the recommendations, but would like to see a new revision of the document.

Olaf will summarize the take-aways of the call on the rfc-interest list.

4. Internet of Things (IOT) Workshop

The board closed the call with a discussion of the proposed March workshop on the Internet of Things, which is also being referred to as the Smart Objects Workshop. Hannes has been coordinating with Jari Arkko (Internet AD) and various WG chairs in this space (core, 6lowpan, roll) to construct such a workshop, and talked with the University in Prague about possibly providing a room.

There was general agreement among IAB members that this is a good area for a workshop and/or other IAB activity, but there was significant debate about whether the IAB has sufficient resources to organize the workshop now. The questions raised were primarily about timing, workload, and schedule deconfliction. For example, when the IAB sponsors a workshop it generally commits to writing a workshop report (in addition to the other organizational tasks like drafting agendas, reviewing papers, etc.). Russ noted that people were already making travel plans for the IETF meeting, so if the workshop is to be co-located with IETF 80 it will be challenging, especially if participation is limited based on paper acceptance. John raised the issue of another similar workshop taking place at the same time. After significant debate, Hannes pledged that if the IAB sponsored the event he would commit the necessary resources to ensure its success. There was rough consensus to proceed, provided the workshop is co-sponsored by the Internet Area.