On March 14, 2014, The National Telecommunications and Information Administration(NTIA) announced its, "intent to transition key Internet domain name functions to the global multistakeholder community."1 This transition is more commonly referred to as the transition of the Internet Assigned Numbers Authority(IANA) functions contract, or sometimes just the IANA contract. This series of three posts will attempt to unwrap exactly what the transition of the IANA contract means, and why it matters to the Internet at large. In the process these posts will investigate what the IANA contract is, its history, the people and policy makers influencing it, and an informed recommendation for its transference.
In this third of three posts on the NTIA IANA transition we'll examine some of the proposals put forth to transition the IANA function. I will then analyze what's exactly at stake, and then put forth my own opinionated proposal for the IANA transition. If you haven't already, check out the first and second posts in the series before reading further.
We have now established the background of the IANA functions contract, its history, its parts and the various parties involved in Internet naming and numbering. Let us now investigate some of the proposals put forth by interested parties, then put forth a proposal of our own.
For a wonderful graphical representation of some of these proposals, InternetNZ, the .nz registry operator created this chart.2
Probably the most widely circulated proposal is what's called the I proposal.3 Pronounced 'eye star', this meta-organization is a coalition of ICANN, all the RIRs, the IETF, the ISOC, and the W3C. It is the who's who of Internet organizations. The name I is in reference to the tendency of the '*' character to act as an inclusive wildcard. It is a way of saying this organization encompasses every Internet organization.
The I* proposal generally changes nothing. It is a proposal for the maintenance of the status-quo. It explicitly states, "The roles of all Internet registry policy bodies stay unchanged." ICANN also makes a commitment to, "provide affirmations to all stakeholders (including governments) that all Internet registry policy bodies and ICANN itself will continue to use open and transparent processes."
The one small exception is to remove Verisign's current role in operationally maintaining the DNS root zone file. I* proposes this role to transition to ICANN.
The Internet Governance Project(IGP), run by Professor Milton Mueller at Syracuse University, has also made a proposal.4 IGP has been a stark critic of ICANN at times, its closed door decision making processes, and its lack of accountability to the larger multistakeholder community.
The main feature of IGP's proposal is to separate the operational tasks of DNS root maintenance with policy considerations. The proposal would create a new entity call the Domain Name System Authority(DNSA) which would handle the operational tasks related to DNS root management. ICANN's role in IP addressing would not change. This produces a contractual relationship whereby ICANN contracts to DNSA to implement its policies. ICANN makes the policy and DNSA carries it out. This proposal would also remove Verisign from their root zone duties. Assigning that duty instead to the DNSA.
The US House of Representatives has also weighed in on the IANA transition. Three separate bills have been introduced to stop the NTIA from relinquishing control of the IANA function.5 All three bills are currently sitting in the Committee on Energy and Commerce, where one would expect them to be combined into a single bill. It remains to be seen what will come of this, if anything.
There have been numerous other ideas floated on how to reorganize the IANA contract. Naturally, most of the discussion centers around what role ICANN should have. There have been proposals to replace the NTIA with another unnamed organization to keep ICANN on an oversight leash. Or to create a new organization strictly for policy decisions and have ICANN keep them on a leash. Almost everyone agrees that Verisign should relinquish its duties in maintaining the DNS root zone. Many proposals were submitted as part of a recent Internet governance meeting in Brazil.6
Analysis and Opinion
This is the section where I, the author, express my opinion with regard to the IANA functions contract transference. These opinions are my own, and do not represent the opinions of my employer, past, present or future.
Instead of picking apart the various proposals this section will take a wider view of naming and numbering in the context of Internet governance. The real issue at stake in the transition of the IANA functions contract is accountability; Who is accountable to whom, and how can we design a governance system that structurally embeds a dependency of accountability.
A question which helps us determine this is; who is the IANA functions holder accountable to, or what do we mean when we say 'the larger Internet community'. The answer is everyone. Every human on the planet has their interests wrapped up in the Internet. Even humans not yet connected to the Internet will eventually be connected. Therefore, every human on the planet should be concerned with the Internet and its governance.
The next question we should ask is then; How do you make an organizational structure accountable to the entire human race? Is there any precedent for this, and if so how did they grapple with it?
We can't consider the United Nations(UN) a good model, nor its predecessor the League of Nations. Both institutions emerged biased by the respective victors of their preceding conflicts. The permanent members of the Security Council are testament to this. In addition, the Internet at it's core rejects the idea of governance through state actors. The NTIA explicitly chose the word 'multistakeholder' in their transition announcement. It's in the first sentence twice, "To support and enhance the multistakeholder model of Internet policymaking and governance, the U.S. Commerce Department’s National Telecommunications and Information Administration (NTIA) today announces its intent to transition key Internet domain name functions to the global multistakeholder community."7
Multistakeholderism means different things to different people, but in this context it means not just state actors. It's an inclusive term that includes states, businesses, civil society and individuals. The IANA functions holder must be accountable to these stakeholders, who are in actuality, the entire human race.
So much depends on the IANA functions holder behaving well, and in the interests of the human race. However, without clear means of accountability, organizations like ICANN eventually stop operating in the interests of the people they represent. Maybe today ICANN has an honest, upstanding staff who endeavor to represent the interests of the Internet community. But history has shown us the limits of unaccountable governance.
Governments formed without structurally embedded systems of accountability are unstable and generally inhumane to those whom they claim to represent. Brutus didn't stab Caesar because he had a problem with Caesar himself, he stabbed him because the Romans had deposed their kings and didn't want new permanent, unaccountable dictators. The authors of the US Constitution understood this as well. No taxation without representation was a rallying cry for accountability.
We don't have to depend entirely on any theoretical possibility of the IANA functions holder acting in bad faith, since we already have plenty of history to show them acting in bad faith. Karl Auerbach was one of the only people ever to be elected to ICANN's board of Board of Directors in 2000. While there, Auerbach asked to see ICANN's corporate records, but was repeatedly rebuffed and stonewalled. He filed suit, and in Auerbach v. ICANN won access to ICANN's corporate records.8 Shortly thereafter, ICANN ended the elected representative positions on their board. This episode details ICANN's lack of willingness to be held accountable, or act in any transparent fashion.
On April 21, 2014 Auerbach wrote a letter to his representative concerning the IANA transition.9 In it he argues that ICANN has always been seeking independence from any accountability, "ICANN is not looking for a change in putative masters; ICANN is looking for independence. [..] ICANN represents a new kind of thing under the sun. The internet is eroding the authority from traditional nation states. The granules of that authority are not disappearing; rather they are aggregating into the new kind of sovereignty that ICANN represents. Unfortunately, that aggregation of authority is not accompanied by any system of accountability except that which ICANN (or rather those who wag ICANN) chooses, voluntarily, to impose upon itself."
Auerbach hits the nail on the head with this. ICANN does not want to be held accountable, is currently investigating ways to avoid accountability, and has a track record of avoiding accountability. We can't expect them to behave better once the NTIA steps out of the process? In contrast, we have every reason to expect ICANN to behave even worse once they no longer have to renew their contract with the NTIA.
According to Auerbach's letter, ICANN generated $400,000,000 in revenue in 2013, yet their role as the IANA steward should be just a clerical one. Much of this money came from fees that ICANN charges DNS registries for new TLD applications. These are limited public resources that ICANN charges rents for.
In the same way that there is only one Mount Rushmore, or Old Faithful Geyser, there is only one .com domain. As such, the same argument that John Muir and others made that caused the creation of the United States national park system can be made here. These are public resources too important for one individual or corporation to control exclusive access to. These resources must be managed in the interests of the human race. Just like the US national park system, the Internet requires new thinking about public and private places. Auerbach is right in that the Internet is eroding the power of nation states, Mueller makes a similar argument10. However, this should not mean public space erodes along with the sovereigns who currently ensure it.
While we don't have to keep private interests from chopping down websites like lumberjacks chop down trees, any limited resource engenders a tussle over control, and DNS names are no different. Since the rise of heavy commercial use of the Internet there have been debates and legal challenges surrounding DNS names. The legal precedent is somewhat murky, especially since trademark law differs by jurisdiction. ICANN has an interest in clarifying this murkiness to avoid endless lawsuits, and provide predictable outcomes for any tussle. On the one hand these efforts can be seen as providing much needed dispute resolution guidelines that help disputing parties resolve conflict. On the other hand, these efforts can be seen as ICANN attempting to become more sovereign, and less accountable to established fora for dispute resolution.
Along with the launch of the new gTLDs in 2013 ICANN launched the Trademark Clearinghouse(TMCH) where trademark owners can register their mark. "The Trademark Clearinghouse is the only authorized and universal means of a brand owner protecting(sic) their trademarks during the launch and take-off periods of every new gTLD."11 Owners who register their trademark at the TMCH get preferential treatment in DNS disputes, and notifications when someone registers a DNS name which may conflict with their trademark.
The TMCH is a clear usurpation of a public function by a private entity for profit. Registering trademarks with the TMCH costs $150 per year with bulk discounts available. This is essentially a tax on trademark owners who care about their brand on the Internet, which is likely akin to all trademark owners everywhere. No doubt there is a need to provide clarity and security to trademark owners on the Internet, and charging a small amount for this is not unreasonable. However, when one organization controls access to this public medium, has no accountability to anyone but themselves, and refuses to operate in a transparent manner, there is clear usurpation of authority. As trademark disputes continue to occur on the Internet, and ICANN continues to evolve as their first arbiter, it will be interesting to see how its position is shaped by court rulings and political pressure.
Currently there is a ripe dispute between the government of Brazil and Amazon.com over who gets the .amazon DNS TLD. Amazon.com claims that since it owns the US trademark on Amazon, it should be given the domain, whereas Brazil claims it has the Amazon river, so it should be given the domain. As of May 2014 the issue has been referred to ICANN's Government Advisory Committee(GAC), which has yet to make a decision.12 Even if ICANN does reach a decision this does not preclude either party from taking either ICANN, or the opposing party to court, as the outcomes of ICANN's extra-judicial rulings can always be further challenged in judicial fora. One might argue that the TMCH provides neither clarity or security given this fact. For ICANN however, the TMCH is simply another revenue stream to be exploited.
We have explored the problems with the IANA status quo and the I* proposal to maintain the status quo. This section will present a proposal to better address the problem of accountability in the IANA functions contract.
The basic proposal is the same as presented by the IGP, but with one significant change, ICANN must become accountable to Internet users. In comparison to today, the IGP proposal removes Verisign and creates a new organization to handle the operations of the DNS root. This new organization is called the DNSA, and is contractually obligated to ICANN for maintaining the operational implementation of ICANN's policies. ICANN is responsible for policy, while the DNSA is responsible for implementing that policy.
In addition to creating the DNSA this proposal would also compel ICANN to reinstate direct participatory representation on the ICANN board of Directors. Currently the board is composed of 16 voting members.13 Eight members appointed by the Nominating Committee(NomCom), two appointed by the Address Supporting Organization(ASO), two appointed by the Country-Code Names Supporting Organization(ccNSO), two appointed by the Generic Names Supporting Organization(GNSO), one at large member elected from the various ICANN user groups around the world, and the president who is elected by the board itself.
The NomCom is composed of six representatives from business, one from the IETF, one from the ASO, one from the ccNSO, and one from civil society. This structure gives an outsized amount of power to the Internet industry, and not the Internet's users. There is only one representative for Internet users on the board, and one representative from civil society on the NomCom.
This proposal would compel ICANN to adopt all but one of its board members from the Internet user community. The only board member not drawn from the user community would be the president. Who would be appointed by the board, as well as sit on the board. The specific method for doing this is less important than the outcome. The method for selecting the current at large member could be expanded to include all members, or a new process could be introduced. The only requirement is that it adhere to participatory democratic principles. If ICANN is going to act like a governing body, its board needs to be elected like a government body.
To those who might argue that introducing accountability, or adversarial process, to ICANN's workings would encumber them or make them inefficient, I say that's the point. The intention of democracy is not to engender expedience or efficiency. The intention of democracy is to engender cooperation, and cooperation takes time.
A large problem in Internet governance is that people don't understand how the Internet works, or the scope of influence of naming and numbering. Including the Internet community in the selection of ICANN's board members would have the additional effect of educating the Internet community. Also, making ICANN more representative of the Internet community would give it more legitimacy in the eyes of governments and the rest of the world.
Another argument against introducing democratic process to the ICANN board composition is that it won't work. This argument says that the world is too large, and no experiment in participatory democracy this large can ever work.
Unfortunately for the purveyors of this claim, ICANN had general elections once in 2000, and they worked. Hans Klein, a Professor in the School of Public Policy at the Georgia Institute of Technology studied ICANN's 2000 election. His conclusion was clear, "The preconditions for democracy were present to some degree, suggesting that global democracy is not impossible – and is even feasible. The creation of a global political community with broad membership did not lead to an explosion of heterogenous issues."14
This is a proposal to transition the IANA function away from the NTIA and into the hands of the global multistakeholder community, not into the hands of ICANN unconditionally. ICANN does not represent the global multistakeholder community, nor should it pretend it does. It can mutate into an organization that does, but to do so it must become accountable to that community. A process it has thus far strongly resisted.
This series of three posts has been an exploration into a very esoteric and complex topic that few people understand, yet that impacts many. Internet naming and numbering are topics that occupy an undersized space in the contemporary zeitgeist relative to their importance.
There isn't a day that goes by without a leading news story about the Internet, but yet so few people understand how its governance functions. This needs to change. Hopefully, this series can play a small part in that educational process.
Milton Mueller, Networks and States:The Global Politics of Internet Governance (Cambridge Massachusetts: MIT Press, 2010) ↩