Wired on Friday: Last week was the inaugural meeting of the Internet Governance Forum. If you don't know what that is, don't worry. It may be fair to say that not even those flying to Athens to attend the forum knew exactly what to expect.

It was billed as a "multi-stakeholder policy dialogue in the field of internet governance". Its aims, power and intent were blurred by diplomatic niceties.

Until last Monday, all you could safely talk about was its history and origins. And those did not bode well at all. The IGF is a creation of the United Nations, the result of the occasionally acrimonious and controversial World Summit on the Information Society (WSIS) held in Tunisia last year.

The pointed question raised at that meeting was: who runs the internet? The traditional answer had been, until recently, "no one". But that answer has left many countries increasingly unsatisfied. Even if the internet had no one actively pushing and pulling its levers of power, it still seemed clear where those levers might be found: in the United States and, in particular, the quasi-independent body created by the US government in 1998, Icann.

But Icann's power is rather papal in nature. While it can get into plenty of trouble when it makes decrees about how the internet should be run, it's not entirely obvious what would happen if it said or did something truly outrageous.

The machines that give Icann its power - the root domain name servers - are scattered across the world. At the cost of splitting the internet, any nation, ISP or even individual could refuse to recognise those machines' authority and pick a new authority to run under. China, to a certain degree, has done this already. While Icann was thrashing around trying to decide how to use non-Western alphabets in domain names, China decided its own standard and went ahead and implemented it. There was nothing much that Icann could do. Icann, like the pope, has no battalions.

It's become clear that many internet groups are as concerned by this lack of power as they are by Icann's arbitrary authority. Topics of conversation in Athens included not so much a call for the US to get its mitts of the internet but for somebody - anybody - to intervene to stop the issues that afflict it.

Civil liberties groups expressed concern that freedom of expression on the internet was being shut down in venues such as China. Representatives mulled the dangers of spam and malware, and bemoaned the lack of consumer protection.

Costs for the poor, and international connectivity costs, begged for some intervention beyond that of the market.

As "father of the internet" Vint Cerf said, the internet's "problem behaviours" called for "co-operative technical, political and legal efforts for their solution".

The unplanned, loose nature of the meeting appeared to work to its advantage. Google felt free to air its concerns over overstrict copyright; human rights groups felt free to publicly criticise China's censorship policies to government representatives.

The meeting prompted a number of semi-spontaneous "dynamic coalitions" - groups that had decided, as though after an enjoyable holiday, to stay in touch and work together.

The issue is less about what Icann and the US is doing but more how much needs to be done and who could step in to take the mantle.

The list of suitors for the role is as large as the papers submitted: from the OECD, to the International Telecommunication Union (ITU), which runs the global phone network, to Wipo (which manages the world's "intellectual property" rules) to national governments.

Perhaps the one that strikes the most fear is the ITU, whose ponderous deliberations have long been the antithesis of the fast and experimental way in which the internet was built (even the ITU's deliberations on the sexy world of internet security include clumsy sentences such as "Cyber space users are very interested in how to enhance protection level of their cyber life and how to prevent harms from various kinds of threats [ sic]").

The question now is how to turn these well-meaning groups into forces that may be able to address these problems, without diverting too much power toward any one of them. Thankfully, it may prove that while the internet has failed to solve the first, its fluidity may prevent any tyranny emerging from the second.

The worst criticism one can lay at events such as the IGF is that they are "talking shops". But, for once, that has probably been what's needed. Separated by their own takes on the internet, companies, governments and non-governmental organisations (NGOs) hadn't considered pooling their efforts to see what points of view they had in common. Some consensus emerged across borders on even the most controversial topics, such as privacy and freedom of expression in Athens.

For now, that's all the United Nations has.

But unlike the real world, consensus is enough to start productive work.

These groups can go back to their respective nations and start hammering out proposals. Indeed, as Icann's fragility has shown, consensus may be the only thing that can be used to glue the whole of the internet together. Build the consensus, and you build a stronger internet. Lose it, and the internet suffers little - you just find yourself out of power and out of luck.

If there's a shift in the levers of power, it's that the consensus-making forum has chosen to move out of the US and into the international community.