As Antonios Broumas has correctly observed, the Internet Governance Forum (IGF) begins life in Athens next week without the means for its participants to agree upon any substantive documents such as resolutions or declarations. Indeed, according to Nitin Desai, the Chairman of its Advisory Group, it is impossible for the IGF to make any decisions, as it “is not a decision-making body. We have no members so we have no power to make decision.” Instead, the only agenda for action that the IGF’s Web site suggests may follow from the first meeting is the emergence of “dynamic coalitions”, which are described as “a group of institutions or people who agree to pursue an initiative started at the inaugural IGF meeting”. However for the IGF to abdicate responsibility for the substantive work of Internet governance in favour of amorphous external coalitions is to reduce the breadth of its mandate. Whilst the closest thing to a constitutional document for the IGF, the Tunis Agenda, does state in paragraph 77 that the IGF is to be a non-binding process, there is a considerable difference between making decisions, and making binding decisions. “Soft-law” documents such as codes, resolutions, declarations, and even model laws, are by nature non-binding. There is therefore nothing to prevent a body that lacks legal authority from developing such documents; if they are well considered and reflect a broad consensus, that body’s stakeholders will tend to adopt them into their own practice, policy or law. If they are not, they will not be. Not only is this commonplace in the development of international (public) and transnational (private) law, it also reflects the way in which Internet standards are developed within the Internet Engineering Task Force (IETF). The Internet Standards Process set out in RFC 2026 provides that an Internet standard is a “specification for which significant implementation and successful operational experience has been obtained” and which “is characterized by a high degree of technical maturity and by a generally held belief that the specified protocol or service provides significant benefit to the Internet community”. Yet it has no binding force; its only force lies in the consensus that has been built up surrounding its suitability as a standard. Although this addresses the objection that the IGF cannot make decisions because its process is required to be non-binding, Nitin Desai in the passage quoted above isolated a second problem in the fact that the IGF “has no members”. This is a question of semantics though, as it could equally be said that the IGF has as many members as wish to participate in it. There are estimated to be over 1000, from government, the private sector, civil society and intergovernmental organisations, attending its first meeting in Athens. But however the IGF’s membership (or lack thereof) is characterised, is Desai correct in identifying it as a problem? If it were a problem, it would be one that equally affected the IETF, as it is structured in the same way. The closest thing to “membership” of the IETF is participation on its mailing lists. And yet the IETF’s experience shows that the lack of a definite membership does not preclude it from making decisions. What it does do is to determine the manner in which those decisions are made. Specifically, it precludes the use of voting, in favour of the use of consensus (“rough consensus”, in the IETF’s case). There is no reason why the IGF could not reach decisions by the same principle. But this raises the final of three problems with the IGF’s capacity to make decisions, and the most valid. As observed by Antonios Broumas in the article mentioned at the beginning of this story, it is simply not structured in such a way as to allow it to do so. There is an Advisory Council, yes (chosen by the UN-appointed Secretariat), but the IGF is otherwise just a series of meetings; four main plenary sessions on the themes of openness, security, diversity and access, and a number of smaller workshops. In comparison, the IETF is divided into eight technical Areas: Applications, General, Internet, Operations and Management, Real-time Applications and Infrastructure, Routing, Security, and Transport. Within each of the Areas are numerous Working Groups established to work on specific projects, usually the development of specifications for a proposed Internet standard. Each Working Group has a Chair, and may have a number of subcommittees known as “design teams” which often perform the bulk of the work in drawing up the specification. Other bodies involved in Internet governance such as the W3C and the ITU, are structured in a similar manner. To facilitate its ability to make decisions, there is no reason (except political) why Working Groups of the IGF could not also be established to focus on particular issue areas, such as for example spam, privacy and interconnection costs. Much as occurs within the IETF, the conclusion of a successful workshop should be a precondition to the formation of such a Working Group, and its formation could be subject to approval of the Advisory Council and ratification of the IGF as a whole. Compare this again to the IGF Secretariat’s conception of the most likely venue for substantive action arising out of the IGF’s first meeting: “dynamic coalitions”. Whilst the IGF may not have a defined membership capable of making decisions by vote, the composition of a dynamic coalition carries far less guarantee of being representative. Additionally, whereas a formal Working Group can be made accountable to the IGF and be required to conduct its affairs with transparency, an informal dynamic coalition is subject to no such safeguards. So of three reasons why it might have been said that the IGF cannot make decisions, only one has been found to be compelling; the lack of a Working Group structure within which for decisions to be made. And that is not so much a reason, as a choice. If there is sufficient will amongst the stakeholders of the IGF that it should have the power to draft soft-law instruments such as resolutions and declarations by consensus, then there is nothing in its mandate that prevents it from doing so. All that stands in the way of an IGF that is effective in developing public policy for the Internet, rather than just discussing it, is the political will needed amongst its stakeholders to create the appropriate Working Group structures needed for it to do so. It follows that the IGF should not be too quick to abdicate its responsibility to unrepresentative and unaccountable “dynamic coalitions”. Instead, it would do well to consider some of the lessons that it could learn from the most successful body that has been engaged in Internet governance over the past twenty years; the IETF.