Controlling Internet Infrastructure, Part I

On March 14, 2014, the United States government announced its intention to end its direct role in overseeing the Internet’s Domain Name System (DNS). The IANA transition, as it is called, is a moment of critical importance in the history of the global network and the relationship between network governance and government control. It is an extraordinarily complex undertaking, both technically and legally, and there is a great deal at stake—but only a small handful of people understand the full scope of the problems involved and can participate intelligently in the public discussion about what entity or system should replace the U.S. government’s role in DNS oversight. It is thus an unfortunate combination of circumstances for informed decision-making and public discussion. This paper seeks to fill at least a part of that gap.

At its core, the Internet’s DNS is a truly remarkable engineering achievement, and its effective functioning has been critical to the spectacular growth of the Internet over the past two decades. In order for the system to work well and to achieve its goal of universal name consistency, the activities of literally millions of individual domain operators must somehow be coordinated. These coordination tasks are conventionally grouped under three separate headings: coordinating the allocation of IP Addresses (the numbers function), coordinating domain name allocation (the naming function), and coordinating protocol development for both the naming and numbering functions. Beginning in the mid-1980s, these tasks were performed by a number of individuals, entities, and institutions—some private, some public, some commercial, some voluntary—under a complicated series of grants and contracts procured and funded by various arms of the U.S. government.

By the mid-1990s, the system started to crack, as increasing public awareness of the value of the DNS (and the Internet in general) led to the emergence of serious questions about a range of DNS policy matters, including the apparent absence of competition in the domain name market, the practice of “cyber-squatting,” and the lack of any formal management structure and accountability mechanisms for an increasingly valuable global resource. In response to these concerns, in 1998 the U.S. government called for the formation of a new, not-for-profit corporation to develop and administer policy for the Internet’s name and address system. Shortly thereafter, the Internet Corporation for Assigned Names and Numbers (ICANN) was born: a California-based non-profit corporation that, with the U.S. government’s blessing, took over control of DNS policy-making through a series of agreements, brokered by the Department of Commerce’s National Telecommunications and Information Administration (NTIA), between the new corporation and the operators of the various components of the DNS.

The ’98-’99 transition was a success; hardly anyone on the Internet noticed when it took place. Yet NTIA’s recognition of the newly-formed corporation was premised on a number of very specific promises that ICANN had made about the way it would be organized, the tasks it would undertake, and the manner in which it would undertake them—including commitments about the structure and selection of the ICANN Board, the implementation of a “Consensus Policy Development Process,” and assurances that ICANN’s role would be limited to a specific, narrow set of issues. To maintain some degree of oversight, NTIA retained control over one vital subset of DNS-related tasks: the so-called “IANA Functions.” The IANA tasks include (1) maintenance of a series of “Internet protocol registries,” (2) allocation of Internet numbering (IP Address) resources, and, (3) maintenance of the authoritative Root Zone File on the “A” Root server and the processing of any modifications to that file. This arrangement gave NTIA substantial leverage over ICANN’s post-transition actions and operations, because re-opening the IANA contract procurement was both a serious and a credible threat to ICANN’s central role in DNS management. Although it is difficult to say exactly how this oversight was exercised, or exactly how it influenced ICANN’s actions, there is little doubt that it served as an effective “backstop” to keep ICANN in line over the past two decades.

The current IANA transition is the logical culmination of the sequence initiated in the 1998-’99 transition, and it presents a significant opportunity for the United States and for the global community of Internet users. Over time, the justifications for a special role for the U.S. government in managing the evolution of the Internet and its governance systems have considerably weakened, as a consequence of both the Internet’s vastly expanding global reach and of questions about the U.S. government’s ability to claim any kind of neutral “stewardship” role for itself with respect to Internet affairs. Particularly in the wake of the 2013 Snowden disclosures, there is considerable evidence that if NTIA had not voluntarily decided to begin the transition, other Internet stakeholders—including important elements of the technical community, foreign governments, and ICANN itself—would have tried to force its hand.

The IANA transition also has important symbolic significance: it is a formal recognition by the United States that the Internet, which the United States government helped usher into existence 30 years ago, is now truly a global public trust. The Internet’s core infrastructure, rather than being the special purview of any one country’s exclusive jurisdiction, needs to evolve in ways that benefit all users, world-wide. And a strong, consensus-based, non-governmental, multi-stakeholder institution at the policy-making center of the DNS is likely to be the best way to ensure that the Internet infrastructure remains free from undue governmental influence. Moreover, getting the transition right has broad implications for the evolution of the Internet governance system. Its success—or failure—could have a significant impact on the shifting dynamics of the global debate more broadly, affecting both the United States’ credibility and the weight of its support for the multistakeholder model.

Yet the risks the transition poses are also high. The DNS is, by design, essentially invisible to the vast majority of Internet users, but if it were to break down, or fragment into multiple competing systems, the impact on Internet use around the world would be substantial. Furthermore, in the wrong hands control over the DNS can be leveraged into control over a much broader universe of Internet activity and communication than that encompassed by the DNS alone. Freed from U.S. government oversight, what is to prevent ICANN from inserting itself into global law-enforcement or governance role far removed from its core commitment to insuring that the DNS runs smoothly and efficiently?

The stakes are high, for everyone who uses the Internet and everyone who is concerned with its future development as a global communications platform. Designing a transition plan that achieves the goal of relinquishing the U.S. government’s oversight over the DNS while eliminating (or at least minimizing) the risks will be a difficult task, one that will require considerably more public attention and debate than it has received up to now. This paper, by explaining the nature of the challenges and the opportunities presented by the transition, lays some of the foundation for that debate, as well as for subsequent papers in this series, in which we will address in greater detail the substance of specific transition proposals now under development, along with our recommendations concerning implementation of what we believe to be the key components of a successful transition process.

Read the full paper here.