Articles

---

Ruling the Root



Review By Bob Heterick

Understanding the Domain Name System controversy
by Milton L. Mueller

Ruling the Root, The MIT Press, $32.95

The Senate Commerce Committee's Subcommittee on Science, Technology and Space held a hearing last week to address widespread dissatisfaction with the organization charged with managing the Internet's address system, the Internet Corporation for Assigned Names and Numbers [ICANN] ... There was little consensus among those who testified.

-- Associated Press, June 18, 2002

Professor Mueller's treatment of the Domain Name System controversy is as current as today's headlines and provides the insights into why the controversy is likely to extend into the next decade. Explanation of the controversy is relatively straightforward -- resolution is anything but.

The first packets were exchanged over ARPAnet just a few days after Neil Armstrong walked on the moon. A long time ago measured in Internet time. The Advanced Research Projects Administration (ARPA) of the Department of Defense was funding research into more robust communications networks. Out of this research came ARPAnet, the TCP/IP protocol suite, NSFnet and the Internet as we know it today. Mueller clearly explains the genesis of the Internet, its transition from a confined research and education network to the commercial phenomena of today and the key players who made it all happen. Along the way he gets the relationships of some of the players a little confused but that doesn't detract from his story.

For an internet to work, each host computer in every connected network must have a unique identifier. As there are likely to be lots of hosts, efficiency suggests a hierarchical structure, assigning each host a number such that the assignment process can be distributed throughout the hierarchy. Long strings of numbers are hard to remember by humans so the use of names arose concurrently -- each of the names being mapped to a unique number identifier.

This hierarchy begins at the root, the top level, where a few domains such as .com, .org and .edu sat at the top of the hierarchy. Who controls the root mattered little so long as everyone accepted that control. For 20 years the emerging, and relatively confined, Internet progressed just fine. It developed its own conflict resolution structures and created a de facto organizational structure. Consensus management of a communications commons by technical experts worked just fine as long as the naming conventions and name and number assignments existed to help hosts find each other.

The emergence of the World Wide Web changed all that. The naming conventions were suddenly being used to identify content, as well as hosts. Content has commercial value and short, easily remembered or guessed names are perceived to have much more value than long strings of numbers or cute, concatenated Star Trek-inspired handles. Names had trademark implications. Names had potential for frequent site visits that translated into increased advertising revenue. In short, the commons was transformed into a commercial marketplace.

The assignment of names and numbers at the top level was a function performed essentially by a single Internet pioneer, Jon Postel, working in concert with the consensus technocracy. The assignment at the second level was handled by a government contractor, Network Solutions Inc. As the number of registrations in the .com domain increased exponentially with the growth of the Web, it was decided to charge for registrations at the second level. This soon became a lucrative business grossing tens of millions of dollars a year for Network Solutions -- yet another aspect of commercialization.

Now the controversy. If there is money to be made at second level registrations why can't I do it in competition with Network Solutions? If I want to create new top level domains -- say .web -- people can only find me if the root puts me in its routing table. If I want to register customers in .com then the root has to divide the unique numbers and adjudicate the uniqueness of names between me and Network Solutions. If I have a well known trademark, like Chevrolet, I don't want the root to be creating new top level domains because it means I have to register Chevrolet in every new domain it creates if I wish to avoid its being appropriated by someone else. Who, in fact, gave all this authority to root in the first place? Who, in fact, owns the Internet and has the right to decide how many and what domain names will exist in the root? Who has the authority to adjudicate conflicts at the second level over names that competing parties wish to own? Mueller develops a host of similar contentious issues that accompany the transition from a consensus commons to a commercial marketplace.

The book focuses on this transition, fully developing the conflicting points of view, the ever-shifting alliances and the attempts by the Internet technical cadre to hang onto control of the root. He chronicles the creation of the Internet Society and the attempts to translate the commons organization to its formal organizational umbrella. And, he brings us to the present with the creation of ICANN.

The Clinton administration, under the leadership of Ira Magaziner, encouraged and promoted the transition of the Internet from an entity more or less controlled by and operated for the United States Government. Internet oversight had transitioned from the Department of Defense to the National Science Foundation and eventually lodging in the Department of Commerce. The effort to complete the transition from the government to private hands culminated with the famous (or infamous, depending on your point of view) White Paper. The White Paper decreed that transition would take place when a private entity was able to garner consensus support of the Internet community for administration of the root. As Mueller points out, there was and there is no consensus sufficient to identify an Internet "community."

Predictably, political coalitions formed to define the required new private entity and the winner was ICANN. ICANN perpetuated the dominance of the old Internet technocracy and absorbed much of the de facto organizational structure of the historic Internet. In the last third of his book, Mueller discusses the consequences of the political compromises necessary to create ICANN. He is particularly critical of the influence of international trademark interests on the development and operation of ICANN. In fact, he appears to have little empathy with either ICANN's governance structure or its pseudo-regulatory role. His criticisms are well developed and clearly presented.

So why are various congressional committees in the United States still debating the role of ICANN and the "ruling of the root?" Because the U.S. government through the Department of Commerce still maintains "ownership" of the Internet. The White Paper's plan to transfer control of the Internet to a new kind of international organization, free from the political push and pull of nation state politics, has failed to materialize.

Like the crazy aunt we keep locked in the basement, everyone is afraid this will get out and we will be bereft of explanations. Absent any of the usual mechanisms (international treaties, transnational regulatory authorities, etc.), ownership of the Internet is forced to lodge with its last clear owner, the United States government. This doesn't sit at all well with our continuing public claims that the Internet is global and is "owned" by the Internet community. Conflicts, rather than being resolved by treaty or other conventional mechanism, simply continue unabated eventually finding their way back to the U.S. Congress where political compromises patch the open wounds but do nothing to treat the underlying disease.

Mueller's criticisms notwithstanding, the Internet has managed to grow, has not degenerated into the anarchy so many feared, and might well be viewed as prospering. Whatever the mistakes and missed opportunities of ICANN, the worst fears of many have not come to pass. The ubiquity of Internet search sites such as Google has mitigated much of the naming convention issues for all but the trademark community. The governance of ICANN is an issue for a very, very small number of people, organizations and states who perceive an economic or political interest in the ruling of the root. Among this group there is little consensus. For the overwhelming mass of Internet users these are not either interesting or important issues.

For those interested in understanding the Domain Name System controversy, Professor Mueller has created a well researched, clearly reasoned and readable book.

Bob Heterick is the former President and CEO of Educom and one-time member of the Federal Networking Council Advisory Committee. Educom was a charter member of the Internet Society and the first CEO of ICANN, Mike Roberts, was a long-time Vice President of Educom.