Internet Governance for IETFers

By: Andrew Sullivan

Date: March 1, 2014

line break image

The Internet Governance Update (IGOVUPDATE) Birds of a Feather (BoF) session was held at IETF 88 in Vancouver, Canada. The topic was unusual in that it was not about a protocol, or even about discussion that might be needed for future technical developments. Instead, this was an effort to encourage understanding by the Internet Engineering Task Force (IETF) of what other actors are doing or planning to do with the standards that the IETF produces. As is normal, the minutes for the BoF are in the IETF meeting proceedings.

There were two main topics of focus for this meeting. In the Web Extensible Internet Registration Data Service (WEIRDS) working group, the IETF is working on a protocol intended to be used instead of WHOIS. The Internet Corporation for Assigned Names and Numbers (ICANN) is also investigating issues related to the data that is available in WHOIS, and how and to whom that data is available. It seemed useful to show how these two efforts affected one another, so half the BoF was devoted to that issue.

The second topic was an overview of what happened at the Internet Governance Forum (IGF) in Bali. This was important partly because the IETF sent some people to the IGF to talk about what the IETF does. In addition, during the IGF some proposals for new Internet governance initiatives emerged.


Chris Disspain presented a description of the progress of the ICANN Expert Working Group (EWG) on gTLD Directory Services. Disspain is the ICANN Board liaison to the EWG. He observed that there are two kinds of issue. One has to do with the protocol itself: the WHOIS protocol has no structure to its data, and is badly designed for the international Internet we have today. The other kind has to do with policy questions such as accuracy, consistency, and privacy.

Next, Murray Kucherawy, WEIRDS Working Group cochair, outlined the progress of the WG in delivering a new Registry Data Access Protocol (RDAP) that uses hypertext transfer protocol (HTTP). The idea is to use the facilities that HTTP already offers in order to reduce the effort required to satisfy all the needs.

The goal of the discussion was to highlight the ways in which external needs of policy-making bodies can be satisfied by IETF work without the IETF work grinding to a halt while those needs are determined. As Olaf Kolkman (the other WEIRDS cochair) observed, the WG is depending on the extensibility of its design. The RDAP specification can proceed for number resources even while the exact problem description for name resources is not ready. Once the latter is ready, more extensions can be added to meet those needs. The result, it is hoped, will be able to satisfy even unknown policy requirements.

Some BoF participants expressed skepticism about this approach. But unless the IETF is prepared to do nothing until external parties establish all requirements, it may not be able to do any work. In the case of WHOIS, the policy disputes have been going on for more than 10 years. If we want to proceed with flexible protocols, this is our way forward.

The Internet Governance Forum

The purpose of the discussion of the IGF was partly to make IETF participants aware of the way in which others in the world view “the technical community,” and partly to begin a discussion of how we might respond to those views. Some of these issues have come to the fore as a result of political developments having little to do with the IETF. Nevertheless, they make the IETF’s job more challenging.

One of the things that came out of the IGF meeting was a plan for a spring 2014 meeting in Brazil on the topic of Internet governance. It is unclear the extent to which this meeting, or others, will continue to favour the multistakeholder approach with which many of us are familiar. A competing model is a more traditional, multilateral approach in which governments negotiate exclusively with one another. Such an approach could be bad for the IETF traditions of standards development.

One of the central issues here is a mismatch of expectations. Organizations that work primarily through or extensively with governments often try to divide issues by representative groups. The IETF does not work that way: we officially take input from individuals, individuals act within working groups, and so on. This can be an awkward interface for government employees, who are often not allowed to deviate from an official government position. At the same time, as a community we have sometimes been loathe to engage in governance topics because we do not have a representative model, and therefore cannot appoint a representative to interact with other bodies.

There was a great deal of feedback after the IGF about how useful it was to have Jari Arkko (as IETF chair) at the meeting, and it seems important to continue this work. In an effort to continue toward a resolution, BoF participants were invited to discuss issues on the [email protected] list.


While the two topics for this meeting may appear to be quite different, they have at their cores the fundamental question of how the IETF can deal with groups involved with policy rather than protocol. The policy/protocol distinction, while often relied upon, is not really as clear as one might wish. As the importance of the Internet grows, and especially in light of changing political circumstances, the IETF needs to establish how best to address policy demands. Advancing that work is what IGOVUPDATE at IETF 88 was all about. There will surely be more of this kind of work at future meetings.