Following a welcome address by IETF chair Russ Housley, who thanked the volunteers and contributors who had made the meeting a success, the IETF 75 administrative plenary kicked into full swing. Danny Aerts, CEO of .SE, which hosted IETF 75 in Stockholm, offered a few comments about the importance of the work being done by the IETF community and the support it receives from the Swedish local Internet community. According to Danny, a primary motivation for having .SE host an IETF meeting was that it provided an opportunity to promote DNS security.
The subject of the Domain Name System Security Extensions (DNSSEC) emerged as one of the key themes of IETF 75 (see panel article, page 12). In his opening address, Russ noted that .SE long ago embraced DNSSEC. Recently, .ORG signed up, which led to all IETF-related domains being signed, including iab.org, iesg.org, ietf.org, and irtf.org. It is expected that the DNS root zone will be signed by the end of the year.
After experimenting with the scheduling of working group (WG) sessions on the Friday afternoon of the IETF meeting, the Internet Engineering Steering Group (IESG) decided to continue that schedule. “The Friday afternoon sessions are necessary,” said Russ. “It helps avoid conflicts, and it provides important face-to-face meeting time.” Russ did, however, recognize that scheduling Friday afternoon WG sessions does conflict with the meeting time usually reserved by the Internet Society Advisory Council.
Jun Murai of the Widely Integrated Distributed Environment (WIDE) Project, which is hosting IETF 76, addressed the plenary audience briefly. Jun invited the audience to come to Hiroshima, Japan, this coming November, pointing attendees to the IETF 76 meeting information that appears on the IETF Web site.
Finally, Russ recognized the efforts of the team that had devised and implemented the newly revised IETF Web site, noting that the new look ended up requiring much more effort than had been originally expected.
Postel Award Announced
Internet Society president and CEO Lynn St. Amour announced that this year's Jonathan B. Postel Service Award recognizes the pioneering work of the four principal investigators who conceived and later led the creation of the Computer Science Network (CSNET). The winners are Peter J. Denning, David Farber, Anthony C. Hearn, and Lawrence Landweber. The award also recognizes Kent Curtis, a U.S. National Science Foundation program officer and visionary who was responsible for encouraging and funding CSNET and for providing the critical bridge that connected the original research undertaken through the Advanced Research Projects Agency Network (ARPANET) to the modern Internet.
Dave Crocker accepted the award in the name of CSNET and announced that the prize money would be donated to charities that support the Internet.
Bob Hinden began his report on the activities of the IETF Administrative Oversight Committee (IAOC) by reflecting on the story of a ship on display at the Vasa museum in Stockholm, where the IETF 75 social was held. As the story goes, the ship sank during its maiden voyage. It is assumed that the cause was the addition of a second layer of cannons, which had not been part of the ship's original design. Bob offered up that story to the IETF community as a cautionary tale, suggesting that it's a good example of what can happen when requirements get added after the specifications are in place.
In financial terms, the IETF is stable. The financial crisis discussed in the past two meetings did not result in a decrease in number of meeting attendees. However, the IAOC is considering introducing day passes for future IETF meetings.
The current year-end forecast assumes a contribution by the Internet Society of USD 1.3 million (USD 1.45 million had been originally budgeted). To date, the IETF has not had to use either Internet Society stimulus money or the shortfall funds of meeting attendees.
Bob further reported that the RFC Editor process has been reviewed and restructured. The new model will include an RFC production centre, an RFC publisher, an editor for the RFC Series, and another editor for independent submissions. The current RFC Editor contract with the University of California's Information Sciences Institute will end at the end of 2009.
The IETF Trust
IETF Trust chair Marshall Eubanks reported that changes have been made to the Trust Legal Provisions (TLP) and the Trust Procedures since the last IETF meeting, including steps to improve dialogue with the community. The Trust intends to solicit community participation on the subject of committees in a variety of ways. In the case of intellectual property rights (IPR) issues, which tend to have long histories, the creation of an IPR and TLP subcommittee was proposed, as was a TLP discussion mailing list.
Currently, the Trust is working on a procedure for future modifications of the TLP. Assuming that consensus on the general ideas can be reached, the Trust would submit in September a formal document describing the new process.
Also with regard to IPR issues, it was noted that there are times when the Trust has to respond to requests for physical proof of IETF attendance. Earlier this year, the Internet Society and the IETF received two requests for such documents, and copies of the relevant material were made and provided. Currently, the blue attendance sheets serve as the only physical proof; everything else is recorded online.
The IETF Trust was created on 15 December 2005 under a Trust agreement between the Corporation for National Research Initiatives and the Internet Society. Unlike the IETF, the Trust is a separate legal entity, and the terms of the Trust agreement cannot easily be changed until July 2010. In advance of that date, the Trust will initiate a review of its activities, and it is planning to have a document ready for review by the community prior to IETF 76.
Other than a few questions directed to the Trust regarding the TLP and other legal activities, the bulk of the open-mic session during the administrative plenary was devoted to the ietf-discuss mailing list. A handful of participants expressed concern that the high volume of traffic on the general discussion mailing list was making it difficult to identify the comments to document Last Calls that get sent to that list. One participant suggested that a separate list be created solely for announcements and discussion of Last Calls. Others felt it would be confusing to add more lists to the existing ones-which include email@example.com and firstname.lastname@example.org that e-mails related to Last Calls can be identified by their subject line and easily filtered from the rest of the traffic.
The general consensus was that it is best to keep the Last Calls on the discussion list. One participant suggested setting up RSS and Atom feeds for each Last Call, which would enable people to subscribe to whichever feed interests them.
IETF 75 Technical Plenary
The Thursday afternoon technical plenary began with Internet Architecture Board (IAB) chair Olaf Kolkman's announcement that the document titled Principles of Internet Host Configuration has been published as RFC 5505 and that Design Choices When Expanding DNS has been published as RFC 5507. The documents titled On RFC Streams Headers and Boilerplates and the RFC Editor Model have both been submitted to the RFC Editor. The IAB is still working on IAB thoughts on IPv6 Network Address Translation, Peer-to-Peer Architectures, Defining the Role and Function of IETF Protocol Parameter Registry Operators, and Evolution of the Internet Protocol model.
The IAB is also still working on Uncoordinated Protocol Development Considered Harmful, a new document that aims to demonstrate the importance of a coordinated approach to successful collaboration between standards development organizations (SDOs) and to explain a model for inter-SDO collaborative protocol development that has been successfully executed by the International Telecommunication Union's Telecommunication Standardization Sector (ITU-T) and the IETF.
During the IAB retreat earlier this year, three items were identified as the main areas of interest: IPv4 and IPv6 coexistence and how to work toward the best results in IPv6 transition, security of the routing data and the routing control plane, and internationalization issues within the DNS and the applications layer and between the DNS and applications.
There were also a number of personnel changes within the community: Eric Burger has been appointed to the Internet Society Board of Trustees, and Patrik FÃ¤ltstrÃ¶m is the new liaison to the ITU-T. Patrik replaces Scott Bradner, who has been serving as liaison since the role was created in 1995.
The IAB responded to the National Telecommunications and Information Administration's notice of inquiry on the upcoming expiration of the joint project agreement with the Internet Corporation for Assigned Names and Numbers, reiterating the role of the IETF with respect to protocol parameters. A link to the response is available here
Network Neutrality Debate
A large part of the IETF 75 technical plenary was devoted to a discussion on the subject of network neutrality. IAB member Marcelo Bagnulo addressed the audience, explaining that the goals of the discussion were threefold: to serve as a means of presenting the debate to the IETF community, to open up a dialogue on how the issue might affect the work of the IETF, and to see whether there is a role for the IETF to play.
Barbara van Schewick, assistant professor of law at Stanford Law School presented an overview of the current debate around network neutrality. The term network neutrality describes a principle whereby a network is free of restrictions on content, sites, or platforms; on the kinds of equipment that may be attached; and on the modes of communication allowed, as well as one whereby communication is not unreasonably degraded by other communication streams.1 Concerns have been raised that broadband providers could use their infrastructures to block Internet applications and content. In the United States in particular, but certainly in other countries as well, the possibility of regulations designed to mandate the neutrality of the Internet has been subject to fierce debate.
Barbara noted that while the network neutrality debate addresses legal content, it does not address the proper treatment of illegal content or illegal applications-which is of particular concern to the IETF-nor does it address interference with Internet use driven by the government. She pointed out that while not all network providers are interested in blocking content, there are incentives to do so, including increased profit, exclusion of unwanted content, and management of network bandwidth.
While the issues associated with network neutrality have become complicated, the key question, according to Barbara, is, Do we want regulation, or is competition sufficient? And what can the IETF do about it? “I have seen people who try to protect nondiscrimination, but how do you do that without being too restrictive?” she asked. “Should all applications be treated alike?
“There are lots of trade-offs,” Barbara pointed out in conclusion. “I have seen that providers will usually argue, “˜If I can't do this, I will make smaller profits, so I will deploy less infrastructure.' The other case is the limiting of network innovation. The choice will mean very different results, and the trade-off needs to be viewed in relation to the choices beforehand.”
Mark Handley, professor of networked systems at University College London, talked about why the IETF should care about the issue of network neutrality. “Much of the debate concerns legal and economic elements, and we are not good at this,” he said. However, the issues related to network neutrality are different in different parts of the world, and the IETF's technologies have to be able to work everywhere in the world.2
What aspects of net neutrality are most relevant to the IETF? Mark said they are the blocking, rate-limiting, and prioritizing of traffic to or from certain destinations or from certain applications.
According to Mark, even though blocking from destinations is not normally an IETF issue, when it comes to security, the IETF does not have a “good story” regarding defence against distributed denial of service or regarding spam prevention. Such a story would likely involve technical mechanisms that are not network neutral. In some places, governments block content because the content is considered illegal or because it poses political problems, which generally are not technical issues. However, the technology is often applied to work around such blocks. Application neutrality, however, is clearly within the scope of the IETF. The question is whether the IETF has provided the right building blocks to allow network operators to manage their networks effectively, taking security and congestion control into account. Mark warned that the community could end up with a network in which innovation can exist only within a narrow set of boundaries or one in which regulators would step in and prohibit broad classes of traffic prioritization.
Mark concluded that network neutrality is, for the most part, an economics problem. To date, the IETF has not given Internet service providers effective tools to make the economics work properly. If this doesn't get fixed, he said, the IETF might be faced with bad legislation and hence architectural stagnation, or with ubiquitous deep packet inspection (DPI) and architectural stagnation. He cautioned the IETF that DPI may become the new network address translation and that if used widely, we would get stuck and unable to change things. This would seriously endanger the openness and innovation of the Internet because DPI describes a mechanism that looks at the actual contents of traffic rather than at just source and destination IP-address and port.
In the discussion following the presentations, Mark emphasized his desire to “identify technical areas in which the IETF can help reduce the net neutrality effects.” He said he does not believe that the IETF should ignore the debate; rather, it should facilitate a discussion about technological possibilities that can help.
According to Ted Hardie, a similar discussion arose at a prior IETF meeting. “How is the discussion about governments or politicians who want to change the network for their benefit different from the discussion about enabling operators to maximize profit?” he asked. “We made a conscious decision to protect the end-to-end network and to allow the network to be controlled by those people who pass the packets rather than the people who initiate the flows. Maybe it's time to let neither profit nor governments control the Net.”
Peter LÃ¶thberg added that if operators simplified their networks and-instead of deploying equipment primarily to maintain its organizational structure-built a highly optimized network with modern technology to simply deliver packets, they would have a reasonable profit, and congestion in the network would be reduced automatically.
When asked to clarify what he meant by his statement that pricing was not an economic challenge but a technological one, Mark responded that at the moment there is no mechanism to hold people accountable for the traffic they are creating. “The bottom line is that we don't even have the technological mechanisms to solve this problem,” he said.
Leslie Daigle said that in her own and the Internet Society's experience in working in both the policy and technical worlds, it has been important to educate regulators and policy makers about the implications of heavy-handed and rigid regulations that focus specifically on current network technologies. “The sweet spot is to get regulators to the point where they understand that the definitions of good behaviour and bad behaviour lie outside the technical realm and in the land of appropriate competition and fairness,” she said.
Another attendee suggested that the IETF look more toward the user-for instance, by providing better authentication mechanisms and “a better means to filter out the bad guys in the middle.”
During the IAB open-mic session, the discussion about what the IETF-and the IAB in particular-can do about network neutrality was continued. Some of the IAB members said they felt that the IAB is not in a position to solve this problem but that it can facilitate the work that is out there. “We create a playing field, not the outcomes,” said Stuart Cheshire.
Dave Oran pointed out that the role of the IAB is to look at the interaction where economics and the protocol architecture for the Internet meet. “Issues of local optima and policy for single networks are not where we add value,” he said.
1. The definition of net neutrality offered here is adapted from a definition found in Wikipedia.
2. For purposes of understanding how to design protocols when users will have differing views, Mark Handley recommends reading “Tussle in Cyberspace: Defining Tomorrow's Internet,” by David Clark, John Wroclawski, Karen Sollins, and Robert Braden. Read more… (PDF).
IAB/IESG Nominating Committee Members 2009-10
Chair: Mary Barnes
- Scott Brim
- David H. Crocker
- Roque Gagliano
- Randall Gellens
- Dorothy Gellert
- Wassim Haddad
- Stephen Kent
- Dimitri Papadimitriou
- Simo Veikkolainen
- Lucy Yong
- Mary Barnes (Chair)
- Joel Halpern (Advisor)
- Henrik Levkowetz (Tools Advisor)v
- Jon Peterson (IAB Liaison)
- Tim Polk (IESG Liaison)
- Henk Uijterwaal (IAOC Liaison)
- Bert Wijnen (ISOC Liaison)