Internet Architecture Board


Report from the IAB before IETF 96

Dear colleagues,

This is the usual IAB report to the community about our activities since the previous meeting (in this case, since IETF 95 in Buenos Aires). As ever, we hope that this form allows you to prepare topics you might want to discuss during the open mic. But of course, if you have views you want to make known by email, we’re easy to reach: send mail to to reach our public discussion list, and to reach just the IAB.

The IAB has a few chartered roles. We confirm the appointments to the IESG and perform standards process oversight and handle appeals. We also perform architectural oversight, we manage the RFC series and the IETF’s relationship with IANA, and we handle liaisons both to ISOC and to other organizations. We try to ensure that anything we do is part of one of these areas of responsibility, and we try to make sure these are all covered.

Here’s what we’ve been doing since IETF 95. You can find mention of each of these on the IAB pages at (where there’s more background, too).

  • We started searching for the new IRTF chair. As we mentioned last time, Lars Eggert decided not to stand for re-appointment this year, so we started the long process of selecting a new chair. IAB members are beginning to interview candidates. (Architectural oversight)
  • Our previously-made but -unannounced appointment to the Internet Society Board of Trustees was announced. From an extremely strong pool of candidates, we selected Richard Barnes. (Liaisons)
  • As part of the IANA transition activities, the IAB sent comments to ICANN on its proposed bylaw changes. We don’t know whether it was because of our comments, but ICANN’s eventually adopted bylaws did change in line with those comments. (IANA oversight)
  • We sent a comment to ISOC on a proposed bylaw change, in order to tidy up some details about how the IETF selects appointees to the ISOC Board of Trustees. The change here required that we clean up our procedures a bit; see below in the document status section. (Liaisons)
  • We held a retreat in Cambridge, MA. More on this below.
  • Also as part of the IANA transition, the IAB started looking for volunteers to appoint to the new ICANN Root Zone Evolution Review Committee (RZERC). Look for an announcement very soon. (IANA oversight, Liaisons)
  • We published RFC 7841, “RFC Streams, Headers, and Boilerplates”. In part, this was in preparation for the RFC Series format changes that are in development. This resulted in moving boilerplate to a website; it now lives at (RFC Series)
  • We sent a reply to a US NTIA request for comments, “The Benefits, Challenges, and Potential Roles for the Government in Fostering the Advancement of the Internet of Things.” (Architectural oversight, Liaisons)
  • We co-sponsored (with Science Foundation Ireland-funded CONNECT Centre) the Internet of Things Software Update Workshop, which ran 13 and 14 June in Dublin, IE. (Architectural oversight)


You can always find the documents the IAB has adopted and is working on at

A big chunk of the documents are currently out to the community for comment. A significant number of those are actually about the RFC format changes, and are very close to publication:

  • draft-iab-html-rfc-03
    HyperText Markup Language Request For Comments Format
  • draft-iab-rfc-css-01
    CSS Requirements for RFCs
  • draft-iab-rfc-framework-06
    RFC Format Framework
  • draft-iab-rfc-nonascii-02
    The Use of Non-ASCII Characters in RFCs
  • draft-iab-rfc-plaintext-03
    Requirements for Plain-Text RFCs
  • draft-iab-rfc-use-of-pdf-02
    PDF for an RFC Series Output Document Format
  • draft-iab-rfcv3-preptool-02
    RFC v3 Prep Tool Description
  • draft-iab-svg-rfc-02
    SVG Drawings for RFCs: SVG 1.2 RFC
  • draft-iab-xml2rfc-04
    The “xml2rfc” version 3 Vocabulary

The other two documents out for comment are a workshop report (draft-iab-carisreport-00, Coordinating Attack Response at Internet Scale (CARIS) Workshop Report), and administrative draft mentioned under our activities, above. (draft-iab-rfc3677bis-00, IETF ISOC Board of Trustee Appointment Procedures).

The IAB has adopted and is working on the following other documents:

  • draft-iab-iotsi-workshop-00
    Report from the Internet of Things (IoT) Semantic Interoperability
    (IOTSI) Workshop 2016
  • draft-iab-web-pki-problems-02
    Problems with the Public Key Infrastructure (PKI) for the World
    Wide Web
  • draft-iab-privsec-confidentiality-mitigations-06
    Confidentiality in the Face of Pervasive Surveillance
  • draft-iab-protocol-transitions-02
    Out With the Old and In With the New: Planning for Protocol

The first is (obviously) a workshop report; we produce one for every IAB workshop. The middle two are products of the Privacy and Security Program; more on programs below. The last of these is a draft that does not fit in any of the programs, because it goes to the heart of the IAB’s responsibility for architectural oversight.


Every year, the IAB holds a retreat, usually not too long after the new IAB is seated. The goal is to try to ensure each IAB member has a clear understanding of what others’ priorities are for the year, and to ensure that we have a common direction so that we work effectively together. This year, we met in Cambridge, Massachusetts, USA, on 17 and 18 May.

Inevitably, quite a bit of the retreat is IAB members talking to each other: the goal, after all, is partly to ensure we’re aligned. But the IAB tries to ensure that the discussions in our retreat are also responsive to factors impinging on the Internet. This year, our topics about those external factors included the ongoing influence of the so-called Internet of Things on the Internet’s architecture; this discussion led directly to the IAB’s comments to the US NTIA in response to their request (see above). We considered whether an IAB program on IoT is something we ought to start. We have not started one yet, but the IAB appeared to be in agreement that one may be needed, if only to identify gaps. More on this below in the discussion of programs.

We spent some time talking about cross-organization workshops: what has worked, what could use improvement, and what more of this we need to do. We had some discussion of trying to find a way to hold a fully-virtual and yet useful workshop. We didn’t come up with an answer, but maybe you have an idea: if so, we’d like to hear about it, because it’s something we’d like to try out.

At the retreat, we also discussed developments in Internet architecture that tend to promote the power or control of the network operator. And we had the good fortune of welcoming Danny Weitzner, Taylor Reynolds, and Dave Clark from MIT’s Internet Policy Research Initiative. Together, they discussed with us the ways that the IAB can and cannot interact effectively with policy makers. One important part of that discussion was the importance of giving policy-makers the kinds of technical analysis that they need without prejudging the policy outcomes.


The IAB organizes its work, for the most part, into programs. There are basically two classes: management programs and architectural programs. The former are how we handle the oversight of various things, and the latter are where we do architectural work. The former are expected to last as long as the IAB continues to have that oversight function; the latter last until the IAB has come to a conclusion on the relevant group of topics, and we expect them to wind down afterwards. Every program on “architectural issues” has a corresponding open list where you can engage with the program. These are all listed on <>.

As we mentioned last time, we have been trying to review all the IAB programs in rotation, in an effort to ensure that they are more successful at keeping work progressing. The success of this has been somewhat mixed. It appears we have been successful in identifying programs that need to be wound down, and we have done that. We have also managed to identify areas where program membership needs some adjustment. We have been, so far, less successful at ensuring that programs that are not producing results begin to be more successful, and somewhat less successful in ensuring that the programs are reviewed in the rotation we originally planned.

Since Buenos Aires, we had a discussion about automatically including Area Directors in programs that impinge on their areas. We decided not to do this automatically for several reasons:

  1. IAB programs, if they’re really working on topics that are directly connected to some active work in some IETF area, should send that work off into the IETF. IAB activities are not some sort of second path to standards work.
  2. There is some concern that, if we make too many formal or strong links between IAB efforts and IETF standards activities, the IAB efforts will become too focussed on the narrow implementation details. Since the IAB’s efforts ought not to be primarily concerned with those kinds of details, it seems better to use more-informal ways of connecting program work to the standardization work in the IETF.
  3. Whenever it makes sense, we can just add the ADs in question to the relevant program anyway, so there is no reason for additional process rules.
  4. ADs already have enough to do, and it’s not clear that more centralization around ADs is better than attracting the interest and attention (and ability to contribute) of people who are working in an area but who are not ADs.

Management programs:

  • IANA Evolution
  • IETF Protocol Registries Oversight Committee (IPROC, with IAOC)

These programs have been attending to the anticipated change to the IANA’s relationship to the US Government. The former of these is responsible to keep track of the IETF’s use of IANA, and the latter ensures that the IANA Memorandum of Understanding with ICANN is administered effectively. Participants in this area have been, perhaps unfortunately, extremely busy as a result of the IANA changes. IPROC was reviewed on 2016-05-25 and IANA oversight was reviewed on 2016-07-06. Once the new Supplemental Agreement with ICANN comes into effect, the IAB’s IPROC program will wind down.

  • Liaison Oversight

Reviewed on 2016-03-23 (i.e. not since IETF 95)

  • RFC Editor (includes RSOC)

This program has of course been at the centre of the RFC series format changes; it was reviewed 2016-06-22.

Architectural issues:

  • Internationalization

There is still reason to be concerned that this topic is in serious trouble around the IETF, but some recent developments suggest some reason for hope. The program was reviewed 2016-04-27. There were some changes to the program membership, and the description changed.

  • IP Stack Evolution

Reviewed 2016-03-02 (i.e. not since IETF 95). This program seems mostly to be generating ideas that then get taken either into the IETF or into the IRTF, rather than producing work itself.

  • Names and Identifiers

The Names and Identifiers program was reviewed on 2016-04-03 (so, strictly speaking at IETF 95, but that happened after the last of these reports). The arcing BoF at IETF 95 was perhaps a little less productive than we might have hoped, though it was nevertheless useful. The program continues to discuss concepts in various drafts, but has not fully adopted any as part of its program of work.

  • Privacy and Security

Reviewed 2016-01-13 (i.e. not since IETF 95).

As noted above, the IAB considered setting up a program for the Internet of Things. One of our doubts about such an effort is that it appears to be a kind of compartmentalization that may not reflect what is needed. Connecting all manner of sensors and actuators to the Internet will have big effects on the world, but it will also have (and is already starting to have) big effects on the way the Internet itself works, in much the way that the introduction of cars to cities meant that the very idea of a city was reshaped. We continue to discuss how we might address this basic issue.


Some of the IAB’s responsibilities, such as personnel appointments, have to be done in closed session. Nevertheless, much of the IAB’s internal wiki contains thing that we think should be open.

The IAB’s wiki apparently originally started as a place to share things that needed to be shared just among the IAB. As a result, it is designed to work in a closed environment. Yet most of the materials there are things that should be public.

We’re working on fixing this. The work started last year with contributions from Mary Barnes and Robert Sparks, and Robert is continuing it. This sort of work is unfortunately somewhat laborious, but it seems likely we’ll have something to report soon.

We’ve also had a couple of requests to consider holding the IAB business meetings in the way that the IESG formal telechats work — accessible to anyone in the community. This would probably result in rearranging how the IAB does some of its work, and it might increase the number of executive sessions — which would therefore never get minuted, so there might actually be a net decrease in transparency. We’re willing to consider that, but before we do it would be helpful to know whether anyone cares. If you think you’d be likely to join IAB business meetings as an observer for those portions that were open, it would be helpful to know it. Please send the IAB your thoughts.


In Buenos Aires, we did not have a technical topic at the plenary. This was the consequence of going to a single plenary in every meeting. The first meeting of the year has more administrative items that have to be completed, so when we decided to go to a single plenary we concluded that we’d lose one technical topic per year; but we intended to have a technical topic at the IETF 96 plenary (indeed, the note in the IETF Journal says to expect one).

We unfortunately had a scheduling and availability snafu this time, and it came late enough that we were unable to recover. We have a “hopper” of excellent plenary topics (which we think need to be informative or entertaining — preferably both — to the general IETF audience), but practical arrangements were hard to make this time. We’re going to address this by starting earlier from now on. We’ve already begun organizing things for IETF 97, so we won’t be disappointed again. We apologize for missing our goal for IETF 96.

Respectfully submitted,
Andrew Sullivan
for the IAB

Originally posted on 2016-07-17