The Unique Political Soul of the IETF

By: Tomas Carlsson

Date: December 7, 2007

line break image

Known to the world are two different political systems. Known to the IETF community is a third system. Whether we call it IETF democracy or Majhum (majority by humming), IETF meeting attendees will know what I mean. Everyone else will have to fight through several levels of abstractions to get a sense of it.

Perhaps the IETF community hasn’t thought of its procedures and processes in terms of a political system. Regardless, I will explain why it is the third form that is relevant. Welcome to the world of mash-up politics.

First, we have democracy. Whether you have it or you don’t have it, you usually want it. In a democracy, decisions are made by representatives who are elected by the voting population. Sometimes the system is referred to as parliamentarism. Whatever you call it, in a democracy, different opinions on a subject are allowed to be expressed prior to a decision’s being made.

Then we have dictatorship. Within this category we also place Muammar al-Gadhafi of Libya and his so-called Third International Theory. In a dictatorship, one person or one group makes decisions without requiring a registered mandate from those affected by the decisions. In most cases, that person or group maintains power with the help of armed forces. Some monarchies operate as dictatorships. In those cases, power is inherited.

In the past, I have studied and written about the procedures of the IETF, but I couldn’t – even in my imagination – have believed it to be so close to direct democracy until I experienced it myself during the IETF meeting in Vancouver.

Politically speaking, the IETF has no equivalent. Its power is intricately tied to the expectation that participants understand the IETF’s rather complex and abstract culture – a culture that, above all, demands that interaction among its participants be handled properly. When accustomed to working among directors, executives, and board members – a culture that embraces a clear organisational hierarchy – one can at first feel annoyed at having to listen to the diverse opinions that are allowed to float up as part of the IETF’s democratic system. I found this to be especially true when the opinions were expressed in the later stages of a process to settle a developed proposal for an In-ternet standard. It is said the IETF has no members and no voting, but in my opinion there are both members and voting – in the same way that an ant is a member of a colony and voting is a means for determining which way the crowd will head next. The IETF may not be a legal entity, but it offers power to the masses and confidence that the rules of interaction will result in the right decisions.

In both ant colonies and the IETF, decisions are made every moment. Small decisions become bigger decisions. They say that practicing democracy is time-consuming. I’d say with regard to the IETF that that is an understatement. If I can identify one single factor that affects the time it takes for a group to move forward in the process, it would be the rigor with which working group chairs demand that participants follow the correct, stated procedures when giving their say.

IETF 70 attendees take a break to chat and catch up on e-mail.
IETF 70 attendees take a break to chat and catch up on e-mail.
Photo Credit: Peter Lötberg, with permission

The word correct is key within the IETF. This is not to say that the goal is for everything that is said to be correct; the goal is to achieve the best technical standards, and within the IETF, doing so means adhering to the rules of engagement and accepted procedures that working group participants have long followed. “This is the correct way of doing it” is an oft-repeated mantra. Correct standards wouldn’t be engineered without the help of an overarching authority. Such authorities used to be Internet veterans with beards that have growth rates that are proportional to the cumulative list of assigned IP numbers. But I have noticed that the fashion has changed lately to one of more bald chins and cheeks. In the same way, over time, the focus changed in the different organisational elements. This is a healthy sign and one that demonstrates a changing reality.

One could ask why it is necessary for the IETF to meet in person three times a year. Isn’t all the hard work that is done in each working group’s mailing list enough? The answer is no. I have seen research on the effectiveness of distance education. The result is that face-to-face interaction is necessary to keep the motivation, the passion, and the understanding among people strong. Face-to-face meetings are also the places people discover the extent to which chemistry translates into well-functioning groups. As ad hoc as they may seem, the personal connection that is found at IETF meetings makes it more likely that these groups will survive.

At first, I found the IETF’s insistence on consensus and the humming as a method to determine rough consensus a bit silly. Eventually, though, the psychological effect grew on me. One can feel the strong hum of a majority in the chest, and no matter how logical your objections, that feeling cannot be erased. It will hold back every not-very-well-grounded opinion. It may not prevent situations where participants are objecting for the sake of objecting, but a good working group chair will in that case make sure the meeting proceeds.

Within the IETF’s system, if I crave the cult status of having initiated, written, and published an IETF standard in the form of a finished RFC, I first have to convince an area director that we need to have a meeting – known as a birds-of-a-feather meeting – to discuss it. Even if I think it is a splendid idea, there will be no working group, no draft, and no nothing if I can’t come up with enough support to keep it going. The best way to get support for your ideas is to first gain respect for your knowledge. You will probably not get that respect in the short time you have at the microphone at the meetings. You earn it in the corridors, or at the late-night get-togethers in the lobby, or in the bar, or on the mailing lists. On the mailing lists in particular, concrete and clever comments and contributions will result in people fighting to hear your opinion.

This is the essence of the third political system: Anyone – no matter their social or cultural background – can take a leadership position within and make a contribution to the IETF system. If you earn respect, if you demonstrate that you are knowledgable, then you will be heard. But it takes time, commitment, and a willingness to participate in a direct democratic system. The entire IETF standards-building process is based on individual contributions that ultimately lead to teamwork. In other words, if you demonstrate wisdom, others will team up around your idea.