Internet Governance – The Clash of Cultures

Draft of talks given in Australia and New Zealand May 2006 – please do not cite or circulate without permission. You can also hear an MP3 of the Canberra version of this talk (slightly abridged) here (MP3, 57MB) thanks to ANU.

Internet Governance – The Clash of Cultures

Danny Butt – http://www.dannybutt.net

Who has heard of John Perry Barlow’s Declaration of the Independence of Cyberspace? An influential text in early web culture, it captures the epic mythology of the new online world, one that was critical in forging a collective sense of possibility in the English-speaking settler nations where web fever was catching hold. Barlow was a Wyoming cattle rancher, and for those of us working in the commercial new media industries the Californian Ideology was a Wired Magazine-sponsored rerun of the Wild West’s escape from the limits of government, and from politics.

“Governments of the Industrial World, you weary giants of flesh and steel, I come from Cyberspace, the new home of Mind. On behalf of the future, I ask you of the past to leave us alone. You are not welcome among us. You have no sovereignty where we gather. […]
Your legal concepts of property, expression, identity, movement, and context do not apply to us. They are all based on matter, and there is no matter here.[…]
You are terrified of your own children, since they are natives in a world where you will always be immigrants.[…]
In our world, whatever the human mind may create can be reproduced and distributed infinitely at no cost. The global conveyance of thought no longer requires your factories to accomplish.”

John Perry Barlow, A Declaration of the Independence of Cyberspace 1996

But the electronic frontier is another story. From today’s vantage point, we can simply note that the anti-immigration provisions of Barlow’s declaration haven’t aged so well. And of course, the “global conveyance of thought” precisely requires the factories and labour of those who are not natives in the digital environment, the bulk of which happens to be located in the Asia Pacific. What would it mean to factor the experience of the Asia Pacific into the bodies that structure Internet governance?

I want to leave you in this talk with three things:

1. Contrary to popular mythology, the Internet is and has always been governed, by both private and governmental actors. As we shall see, one government in particular plays a pivotal role.
2. The complaints about the existing regimes of Internet governance are substantive. There are many areas where it fails to meet its own principles of openness.
3. The debate over Internet governance reflects long-held differences in culture and power between the US and its allies, and other parts of the world.

Stories

Let me start with a great story by David Maher, a trademark lawyer who was involved in some of the early discussions behind the setting up of ICANN, the organisation which controls Internet Resources such as domain names and IP addresses. Maher describes going to a “Summit on the Internet” meeting that was hosted by the Internet Society in the early 90s. Maher was there to discuss the problems that the domain name system was facing with cybersquatting and issues around trademark law. Jon Postel, the engineer who was then single-handedly responsible for deciding who could have what domain name, did his best to discourage Maher from speaking, suggesting that Maher could “sit in the audience rather than on the podium, and say something when the others had finished”. The thought of sharing a stage with a “lawyer” was abhorrent to this group, as Maher found out when he was told, “if he wore a suit” to a follow up meeting that “no one would talk to him”. When Maher suggested to the summit that the way that domain names were being handled was on a collision course with trademark law, Vint Cerf, the Chair of ICANN and “grandfather of the Internet”, was apoplectic, saying “How dare you tell us how to run the Internet”. A few years later, ICANN would be working with the World Intellectual Property Organisation to come up with a set of rules about how intellectual property interests would be protected online, so I guess they got over trying to run it themselves.
The reason stories like this are interesting to me is because the world of Internet governance is dominated by engineers, and engineers tend to talk about systems rather than people. Now don’t get me wrong, both systems and engineers are very important. I’ve worked in the new media sector most of my life, and I know intimately how a bad platform or software choice can cause all kinds of grief. However, there is a tendency among systems people to think that success is merely a matter of getting the specifications for the systems right. But when it comes to governance, even on the Internet, we’re talking about people, and people don’t work like systems, and can’t be tested and guaranteed like systems. So the qualities that make a piece of technology function don’t always translate into the real world of humans.

Geeks tend to have a belief in systems that is matched only by neo-classical economists, philosophers and some of the more hardline branches of Marxist political theory. You get the feeling sometimes that the engineers could run a great Internet if it didn’t have all these stupid people on it. That attitude, which is embodied in Maher’s encounter with Cerf, also permeates many of the forums where Internet governance decisions are made. In the Internet world a lot of emphasis gets put on the formal rules of any process, more than what actually happens, and any problems are usually about “people not understanding the systems properly.” It’s the difference between a discussion of democracy that insists that “anyone can be a leader” versus one that discusses how Bush Jr following his Dad into office might not have been a fluke 1 in 200 million chance.

I’ve been involved in Internet projects since 1993, but like most people I’ve only paid partial attention to how the whole thing is organised. This is probably because, as English-speaking person, the setup has worked pretty well for me. But for the last little while I’ve been an anthropologist in the world of international internet governance, and I’ve tried to adopt the perspective of the proverbial alien from outer space – looking at what happens and in particular the gap between the way people talk about power and where power actually lies. What I’ve found is that the world of Internet Governance is a fascinating place filled with very smart, usually well-intentioned people who have no idea how unusual they are in the global scheme of things. And somehow, they ended up being left at the controls of some of the most important infrastructure in the contemporary world.

What is governance?

What do we mean by governance? Often people think of it as being synonymous with “government”, but it’s more than that. Governance can be defined as comprising the traditions, institutions and processes that determine how power is exercised, how stakeholders are given a voice, and how decisions are made. This can be in the private sector (“corporate governance”) or the public sector. Note that there are formal and informal aspects to governance – the gap between these will be a theme for the talk.
One of the biggest myths about the Internet, as my colleague Ang Peng Hwa puts it in the book, is that it isn’t governed. Technically, you can say that the Internet is coordinated rather than governed, and it is true that there is no one place where you can go to and say – there’s the Internet. But in reality, there are a number of different areas where self or state regulation is in place, and these are areas where the analysis of governance is useful.

Three areas of governance

The UN Working Group on Internet Governance differentiated clusters of issues related to Internet Governance – there are a lot of them and they’re outlined in the book. I’d summarise them as 1) access issues, 2) issues related to use of the Internet, and 3) issues to do with coordination of Internet resources.

Access

The most basic issues relate to access: what access can you get, how much does it cost? This will be the result of market transactions between whoever owns the pipes. Unlike the phone system, which was mostly set up by governments, there was no provision in data carriage for charging models that would encourage extension of the network in less financially rewarding areas. Of course, another thing to remember is that the US has been very active for decades in trying to overturn the settlement regime where countries receive money for connecting calls from elsewhere.

Here’s a map that outlines some main routes on the North American Internet. You can see that there’s no one network, but a few big private players and a number of very small ones. So the network is not ungoverned – the decisions of those large players have large impacts, and all their development is situated within a legal regime that determines what is possible to build, how competition works, what kinds of contracts are enforceable, etc. However, if the major players make a decision that is not in the public interest, there’s no one place you can go to sort out the system.

This is one area where we see a clear gap between how things should operate in an ideal world and how they actually operate. The theory is that the Internet, as a network of networks sharing a common protocol, almost self-organises to find the most efficient way of sending information from one computer to another. The cost involved should be in connecting to the network. Then, if you run one of the large routing points that’s connected to a lot of others, you get some traffic coming to your network, you get some traffic leaving your network, and it all sorts itself out – swings and roundabouts. In practice, large networks have the power to set the pricing agenda for connectivity, and increasingly they refuse to carry the traffic of other networks as a matter of course. This can mean, for example, that traffic between networks in New Zealand might be routed via the United States because there isn’t a peering agreement between the two networks. The bottom line is that the large networks call the shots. “Net Neutrality” is probably worth a paper on its own, but it’s enough to say that although there has been an ideology about the internet as “neutral”, it never has been (although the protocols might be). The internet exists mostly in the private sector, and the private sector’s job is primarily to accumulate capital by excluding others from its property to create a desirable product/service. There are strong economic incentives for pipe-owners to find the best ways of monetising their investment, which may or may not include open-ness.

Internet Use

Then, there are the issues related to Internet use: spam, cybercrime, intellectual property, etc. These are extremely complex but obviously critical. It is in these areas where the gaps in Internet Governance become evident. For better and worse there is no way of systematically coordinating the entire system to address these issues. These issues are transforming the Internet and being transformed by the Internet, but they are not limited to it. So you can’t necessarily invent Internet-only processes for dealing with issues like cybercrime – they have to be harmonised with existing bodies that have responsibility for them. It’s here also, that the limitations of the private/consensus model of Internet governance are evident. Take for example, the history of Intellectual Property discussions with respect to domain names, which in the Internet arena go something like this:

1. These issues are not important in the Internet (remember Barlow and Cerf’s comments)
2. OK, they might be important, we’ll get a few basic processes in place to deal with the issues of the largest players, but essentially we’re not touching it because it’s not really important.
3. Hmm, it appears that there are massive financial and legal implications to these decisions. We’ll work with organisations like WIPO to implement a new dispute procedure, which happens to be a bit friendlier to US business interests than earlier IP coordination mechanisms.

Coordination of Resources

Finally there is the area of resource coordination. This is what most people think about when it comes to Internet Governance, and it is a byzantine, acronym filled world of domain names, IP addresses, root servers, and protocols. I think the ironic thing about the false “UN vs. ICANN” opposition is how much both sides love acronyms and have processes that are almost impenetrable for the average user. Most of the governance discussion is about this area because it is where there are identifiable power bases that can be analysed. Although the computers on the network are decentralised, there are some areas of central coordination around domain names, IP addresses, protocols, and servers.

ICANN: IP addresses and DNS resources

As you probably know, the Internet is not a centralised network, but a network of networks that communicates using TCP/IP, the protocols set by Vint Cerf and Bob Kahn back in the 70s. The rapid growth of the Internet caused issues in keeping track of which computers were on it. A group including Jon Postel and others created the domain name system (DNS) to make Internet navigation easier. With DNS, users can type host names instead of numbers like “106.27.3.72.” In 1984 RFC 920 established 7 generic “top level domains” (gTLDs, including .com, .net, .org and .gov) to provide domain space for corporations, non-profits, schools, networks, US government offices and the US military. This would be a decision with profound and unforseen commercial implications. As Wolfgang Kleinwatcher has observed, the creation of a new domain is creating a new domain is like the creation of “new territory in cyberspace” and has unavoidable economic, political, and cultural implications. At the time of these developments certain assumptions would be made about “the user” which would have long-running consequences as the governance issues developed.

The principle of delegation says that whoever has responsibility for a particular domain has delegated authority to control all beneath it. The creation of just seven domains places an inordinate amount of power in whoever controls .com. As the Internet grew rapidly, this would become a license to print money. The controls on what gTLDs can exist has become one of the major sticking points in Internet Governance, and the call for the creation of new domains through a transparent process remains one of the loudest complaints levelled at ICANN.

It is worth remembering that ICANN do not control the protocols of Internet transmission – but simply the allocation of IP addresses and the domain name system. These can be seen as equivalent to control of the phone numbers and the phone book, rather than the phone system. There aren’t too many issues related to TCP-IP, or the protocols that underpin the Internet. And even IP addressing, while more controversial, appears relatively stable under the RIRs such as APNIC based here in Brisbane. But when it comes to limited resources, there is money to be made, and where there is money to be made, there is political struggle.

People in the technical community have a habit of downplaying the political issues at stake here. In a famous comment, Esther Dyson, former chair of ICANN and noted Internet personality, could say in 1998:

ICANN does not “aspire to address” any Internet governance issues; in effect, it governs the plumbing, not the people. It has a very limited mandate to administer certain (largely technical) aspects of the Internet infrastructure in general and the Domain Name System in particular.

However, as we have seen, the relationship between the plumbing and the people is the very reason the domain name system exists. And shortly after Dyson’s comments, Verisign would buy Network Solutions in 2000 for $US 21billion on the basis of a business that would be instantly rendered much less valuable if the restriction on new gTLDs was lifted. Whenever there are large amounts of money around, there is governance. Especially when one government is a particular player.

Role of the US Government

The period since 1987 sees the devolution of the DNS to control by the US-dominated private sector, with the US Government maintaining a long leash that it could pull whenever control threatened to relocate to international forums. The most notable example comes in the period of crisis around the performance of Network Solutions, the company with an effective monopoly on registrations in the .com domain space. An unlikely coalition of Internet old-timers in The Internet Society and IANA, who were unhappy with the monopoly capitalism brought about through privatisation, paired with representatives from international organisations such as the ITU and WIPO to form the Internet International Ad Hoc Committee (IAHC) in 96-97. They proposed policy and procedure changes for administering gTLDs – a gTLD-MoU as the proposal became known – in particular seeking to strengthen the role of self-identified stakeholders who felt shut out of the existing processes which seemed designed to bolster the registry coffers more than anything else. The IAHC included trademark people and Internet people (a remarkable turnaround from the gap David Maher mentioned between these groups just a few years earlier), as well as privacy advocates, standards bodies and Internet and telecommunications carriers.

The opponents of this plan, particularly US-based business interests, made themselves active in the many US govt Committee hearings on the subject at the time. Ross Rader notes “perhaps most damaging to the gTLD-MoU was the comments of Sernovitz from the Association for Interactive Media (AIM), who denounced the IAHC and IANA as ‘betraying the United States to the governments of Libya and Iraq.’ While the claims were never completely substantiated or formally investigated, they damaged the credibility of the group behind the gTLD-MoU [in the eyes of the US Government]. Further, the total of all the comments successfully managed to call into focus the potential shift of power that the Internet represented to a country other than the United States.”
In November 1998 the US Department of Commerce signed a Memorandum of Understanding (MoU) with ICANN, a non-profit based in California, delegating it the authority for the apportioning of critical Internet resources – domain names and IP addresses. This MoU still exists – effective control over ICANN is held by the US Department of Commerce. Given USG’s history of resistance to internationalisation, you can understand why numerous governmental and non-governmental actors from outside the US might find it troubling that their use of the Internet is dependent on an entity functioning under a US govt agency. Particularly when that agency was birthed under political pressures that scuttled broad, multistakeholder management practices with unsubstantiated fear mongering about foreign governments.

Domains and their control

The most controversial issues in Internet Governance lie in areas under ICANN’s control. These are
1) Control of country-code Top Level Domains (ccTLDs)
2) The process for creation and approval of generic top-level domains;
3) The development of Internationalised domain names (IDNs).
These are the three issues that I think are most likely to give ICANN significant trouble as they move forward.

Country Code Top Level Domains

Initially, the process of keeping track of domain names and who owned them is a task that Jon Postel did by himself, reportedly in a paper notebook, and it’s now under the control of ICANN. Here is where you get some real action in the governance stakes. Postel originally decided that anyone who was “trustworthy” could administer a country code domain, whether or not they were associated with that country’s government (in fact, in geek culture, being from government might count against you being thought of as trustworthy).

When the Internet changed from a marginal research network to critical economic infrastructure during the 90s, governments understandably thought that they should have some role to play in the management of their country’s identity, because they were responsible for the maintenance of their country’s infrastructure in every other field. In some cases these could be worked out, such as in Australia.

But, for example, countries like Niue have never had control of their domain, and the story is instructive. In 1997, a former US Peace Corp working for the Niuean government set up a business arrangement with an American business man and approached the Niue government to control < .nu> under the name of the Internet Users Society of Niue (IUSN). The IUSN (a non-profitable company), then re-delegated the management and marketing responsibilities to the US-based company, NU Domain Ltd. This is the company that realized the revenue earning capacity of < .nu>, especially by marketing it to Scandinavian countries, where it carries connotations of words such as “brand new”. This struggle is ongoing and is but one unforseen example of a faith in decentralised private contracts not quite delivering the public policy benefits intended.

Top Level Domains

The rationale for limitation on generic top-level domains is that it is unclear how many new domains the Domain Name System could handle. Unfortunately, the level of technical opinion varies greatly on this issue. One thing that is known is that John Postel declared that another fifty could be added without causing significant additional stress. Why haven’t they been? “Stability and security” is the stated reason, but one must also wonder at how much pressure is asserted by incumbents who do well out of the situation where the number of domains is limited. The process for approval of new gTLDs – a very poorly documented one by most accounts – is now being extended to ensure that proposed new top level domains are accompanied by a business plan to show financial sustainability for whoever successfully bids for them. Unfortunately, as ICANN committee member Bret Fausett noted, ICANN can’t agree on criteria to assess the business plans. Most of the ones they’ve approved so far (.aero, .travel, .museum) have not been huge successes. Further, one they are looking to approve which would be widely taken up (.xxx) has foundered on political issues.

IDNs

The impenetrable mechanisms for the approval of new TLDs like .travel becomes even more bizarre if you think that there are an estimated 60 million out of the 100million internet users in China who browse the web *only* in Chinese, but it is so far not possible to have the Chinese language equivalent of .com. This lack of progress on internationalised domain names, or IDNs, is perhaps the largest challenge to the established governance order. Now as you’ll remember, the domain name system is not to make computers work, but is to make it easier for humans to access the various computers. Unfortunately, at the time the domain name system was invented, no one ever thought you might have humans working in languages other than English. That’s not surprising for an overwhelmingly US-based research network. The bigger question is about what’s happened over the last decade as hundreds of millions of people who don’t speak English have come online. The answer, unfortunately, is not much.

Initially, it was no real problem to use a roman script for domain names and expect others to do the same. As the Internet expanded, it was recognised that it would be good to use other character scripts in domain names, but the DNS only works with roman script, and it was decided that upgrading it would be too disruptive. Frightened by the appearance of keyword systems that did DNS-like services in Korea and elsewhere, there was a move to find a solution that was backwards compatible with the existing system. The IDN system developed under the auspices of the IETF translates non-roman scripts into roman script and back again. As it turns out, there are all kinds of intellectual property and usability issues (for example – if you register mcdonalds.com in English do you have to do it in every script?) and influential technologists such as John Klensin are beginning to wonder whether the DNS isn’t the best way of dealing with all this. Which would be fine – except that the DNS is all we have at the moment and the decisions that are made about it are still dominated by the interests of those in English-speaking countries. And those established interests are not really that keen on proposing alternative systems to the DNS, which currently makes them a lot of money. It was not until the end of 2005 that ICANN committed itself to serious work on the issue, and there is a long, long way to go.

Let’s turn this around and ask it in a different way. What if the network had originated in Japan, and for the first years of its existence you could only use Japanese? Then, as the network expands, you have a system that translates some English words into Japanese but forces you to use strange Japanese English idioms while it appears that the Japanese can do whatever they want – particularly when DNS became a branding platform in 2000. The system would seem to assist companies who had Japanese brands (who, funnily enough, mostly happened to be in Japan). When you complain they tell you that there are all these mailing lists, where business is aggressively conducted in various Japanese dialects, where you are welcome to make your thoughts known. I think the response among the US and friends would be to get moving on an alternate system as fast as possible. Particularly if the chair of the standards-setting body said “There is no getting away from the unique root of the public DNS.” (i.e., don’t think about setting up a different system that would work for you).

The threat of alternate roots

This is the largest implication of the current contests over Internet Governance – the threat of alternate roots and domain name systems. At the moment, purely by agreement, almost everyone uses the same phone book that is distributed around the world’s root servers. If some significant figures – say, hypothetically, large ISPs in China – became fed up with the ICANN phone book and wanted to use a different one that allowed Chinese characters, there would be nothing stopping them from doing so. They would probably copy the existing list of domain names and their associated addresses, and then add some extra lines to it for Chinese domains they’ve registered without waiting for ICANN/IANA approval. If you did a lot of business in China, this new “unofficial” phone book would be a lot more useful than the official one, even if it was only 99% accurate.

Some problems could emerge in this scenario. To give one example – what happens if there are conflicting IP addresses for a single name between the new system and the existing system? This could easily happen when a new venture sets up in China and builds an identity around a particular domain name that it has purchased locally – but then finds that a company in Singapore is using the same name for its website. For the user, you would not be able to tell in advance which site you were going to when you type in the domain name, because it will depend on which DNS system your ISP is using, which is not always transparent.

It’s for these reasons that those involved in Internet Governance (and ICANN in particular) seek to maintain a single and unique root zone file.

Personally, I believe the threat of alternate roots is overrated. There are Internet messaging systems like Skype and AIM that don’t rely at all on the domain name system, and finding other ways of handling the problem is possible. And let’s remember, that international business continues to take place even though not everyone speaks the same language. It’s definitely a good thing to feel like you can access any computer, anywhere. But if the cost of that globalness is being bound to a system that may not be in your interests, as it is for a lot of people, then it suddenly becomes less appealing.

It’s a delicate political game going on. No one wants to split the root. But if it came down to a game of chicken, where you were forced to choose between a domain name system which was “official”, or one which was pretty close to the official one but had an extra 30 000 Chinese-language websites in it, which one would you choose? This is the impetus for Internet Governance reform – the alternative is that economics may end up driving the decision and it won’t be in favour of the existing players.

Governance methods

I want to finish by evaluating how these tensions are occurring and how they relate to other governance issues more generally.

Principles and mechanisms

In their excellent book Global Business Regulation, John Braithwaite and Peter Drahos suggest that the key factors that impact upon regulatory regimes can be separated into actors, principles, and mechanisms. Principles set the high-level agenda for governance, while mechanisms are the low-level processes by which things work. So for example, both Australia and New Zealand are described as governed on the democratic principle. But the mechanisms by which the principle is made real – for example, Australia’s preferential voting system versus NZ’s proportional representation system – mean that the activity of people participating in that democracy can be quite different. Under proportional representation a minor party is a much more viable parliamentary option, and so strategies of political parties and voters are different under these different mechanisms.

Contests happen over both principles and mechanisms. In small groups, it is easy to resolve conflict directly, so complex mechanisms aren’t needed, and much of the dialogue is about principles that bring the group together. When a small group of engineers decided to set up a way of linking computer networks together to form the Internet, they spent a lot of time thinking about the principles of openness, decentralisation, and independence that would go into the network. While some mechanisms for how these would be developed weren’t so sophisticated, it didn’t really matter that much because, as Maher’s story showed us, if you were committed you could get involved with the policy mechanisms even if people tried to dissuade you.

The problems with existing Internet Governance processes largely stem from thinking that their principles are sufficient for effective governance. They might have been at one stage. But now, everyone from your Latin American activist-Linux developer to Microsoft is in favour of “empowering the user” and “openness”. When the French and the US argue over “freedom”, who’s right? The way these principles are operationalised is a political contest among people with very different stakes and capacities. And one’s orientation toward a particular body or group depends a lot on your history with them.

Consensus and openness

Speaking of the French, they have a great saying: “Those who don’t do politics, get done in by politics”.

Internet governance organisations, as befits their engineering heritage, seek to avoid the concept of the political at any opportunity. Internet Governance began with volunteers and research organizations that created authority through consensus. According to John Palfrey, as we have seen, many of those active in ICANN today, particularly in the more technical capacities, come from the IETF tradition of “rough consensus” sought by “listening to the hums” in the room. Generally, consensus works in relatively homogenous environments where there is a shared understanding of what success will look like. That applies to a group of software engineers discussing technical protocols, but it definitely does not apply to the global base of Internet users and the Internet industry. Because consensus as a mechanism is weak, dissenting voices have never fared as well on mailing lists as established interests. However, any questioning of this mechanism results in a fierce defence of “openness, democracy, and consensus.”
For example, M. Stuart Lynn, upon appointment as president and chief executive, said, “ICANN takes its lead not from me but from the Internet community as a whole.” The 100 million Internet users in China who currently have little direct input into Internet Governance processes might treat this argument with some scepticism.

Part of the rationale for governance regimes in the commercial sector is that there are people and groups outside your management team who are critical to your organisation’s development. Effective governance processes ensure that an organisation’s strategic direction is responsive to those people and groups. In the case of Enron, we saw the dangers of groupthink, of believing that the people around the table at any particular time will make the best possible decision. This is the logic of consensus, which is predicated on the existence of a shared culture.
However, in more complex environments, the issues are somewhat different. Business school advice for CEOs now focuses less on building a shared culture, and more on remaining open to different and challenging views. Because you might be out of touch with your sources of future growth without you even knowing it. Then the game changes and you find that someone else is where you once were. IBM failing to keep up with the PC boom in the 1980s is a commonly quoted example.

The key issue with the Internet Governance regimes, which I think reflects a broader divide between approaches to governance, is their orientation toward diversity and conflict. The current mode of ICANN is one of defensiveness. If there is criticism of a process, it will immediately be met by a flurry of defences of its working principles and openness and innovation. After a century of psychoanalysis we know that defensiveness is essentially an expression of fear and resistance to change. There’s no upside to it – while it temporarily makes you feel better, your critics will smell the fear and come after you even harder. The only way to remove tension is to turn toward it and have a dialogue with it, allow it to change you.

To create forums where not only diverse, but also sometimes opposed perspectives can be brought into productive conversation is the great challenge of governance. In this respect, I would encourage people to make a close examination of the intergovernmental system and to realise that its agencies have been remarkably successful at this in various times. Of course, such a system is not likely to always be “efficient”, or “flexible”, or “innovative and dynamic” – particularly when compared to a smaller, less diverse private sector organisation. But then, as the recent history of privatisation in public utilities has shown us, focusing on principles of efficient flexibility might come with its own downsides in the areas of universal service provision, accountability to smaller stakeholders, and even “stability and security”. Personally, I’m less keen for my electricity company to be “flexible and innovative” and more keen for it to ensure I’ll receive power.

As a geek and a theorist, I love principles. But as a businessperson and consultant, I am usually drawn back to reality, and have the role of reminding organisations what is or isn’t within their capabilities in real life. I am especially sensitive to when people contrast real life to principles and vice versa. It seems to me that when the established players in Internet Governance are criticised for their real-life lack of diversity and responsiveness, they respond with principles. When confronted with issues that have been dealt with for many years other bodies, such as intellectual property or international cooperation, they assert that the Internet is different and that it needs to be dealt with according to their principles.
But the principles aren’t what matters – what matters is that in the relentless focus on the novelty of Internet governance, we risk disrespecting the work that has been done in the past on all sorts of relevant global issues such as shipping, environmental change. If we ignore the people who have undertaken that work, we will fail to use their experience to develop our capacity to deal with complex governance situations.

It is possible to believe, if you spend a lot of time in Internet culture, that governments are bad, that geography doesn’t matter, that any space for “individual participation” is the best measure of openness. For the anthropologist in the house of Internet Governance, the first thing you notice is that the history of Internet development has actually been all about one government, that the power bases are not at all global but concentrated in Europe and the US, and that most of the individuals on the Internet have no way in hell of participating in any of the decisions about how it’s run. It’s quite a disconnect.

That’s a problem of legitimacy, although as we know legitimacy isn’t a prerequisite for holding power. I think the bigger problem is that the dominant ideology of Internet Governance is drawn from a very simplistic view of people and culture that is visibly based in the culture of English-speaking settler nations, and these no longer make up the majority of Internet users. Personally, I would love to see truly global, truly user-driven mechanisms for the management of the Internet. But I think that to get there we need to develop sophisticated and relatively neutral mechanisms for mediating differences, and, much as we did in the post-dotcom era, to remember that some of the things people have learnt over the years in the offline world might be useful to us.