6.4 Conflicts over the Root | Ruling the Root: Internet Governance and the Taming of Cyberspace

< Day Day Up >

In the DNS hierarchy, the power to add new top-level domains or to assign existing top-level names to specific applicants is held by whoever defines the root zone file. But where did the formal authority for this lie? Who owned the name and address spaces? More than a year before the big push to create new top-level domains, the privatization, commercialization, and internationalization of the Internet had prompted discussions of this question in the technical community and the U.S. government. Postel himself, in an October 1994 report on the problems caused by the rapid growth of domain name registration, admitted that the bigger problem underlying it all was that 'it is unclear who actually controls the name space and what is fair procedure.' ^{[45 ]}The commercialization of domain names made this question more difficult to answer, by raising the stakes and bringing new interest groups into the dialogue.

Between 1994 and 1996 three distinct parties emerged to assert claims on the root: the Internet Society (ISOC), the U.S. government, and alternative root servers.

6.4.1 The Internet Society Claims the Root

As noted in chapter 5, the Internet engineering community had created its own authority structure composed of the Internet Engineering Task Force (IETF), IANA, and the Internet Architecture Board. Superimposed over this structure, rather loosely and tenuously at this point, was the Internet Society; its purpose was to provide a corporate identity, legal protection, and financial support to the other components when needed. The Internet Architecture Board already had been fully incorporated into the Internet Society. The IETF rank and file, however, did not yet identify with ISOC. The relationship between them was a 'cantankerous' one, with doubts still being openly voiced about what ISOC was and whether it was of any benefit to the community. ^{[46 ]}

Just before the domain name wars erupted, the IAB and the Internet Society were attempting to transfer formal authority over the root into ISOC's fledgling organizational structure. In July 1994, Postel prepared a draft charter for IANA, proposing that the Internet Society's board of trustees would delegate to the IAB the right to select the IANA. Although the model for 'chartering' IANA was IAB's movement under ISOC's umbrella in 1992, IANA's situation was more complicated. IANA was not an informally constituted committee but a set of functions performed pursuant to government contracts with ISI. The name and address spaces could be considered valuable resources. In effect, Postel was proposing that these functions and resources be privatized. A final draft of the proposal, circulated in February 1995, encountered resistance from parties in the Federal Networking Council (see section 6.4.2). The controversies over charging and new top-level domains intervened before those issues could be resolved.

Draft-postel, drawn up only a few months later, was shaped to a significant degree by the desire to operationalize that new role. In June 1996, at its annual meeting in Montreal, the Internet Society's board of trustees voted in principle to support the proposal. The Internet Society was now formally backing a plan to assign commercially valuable property rights in top-level domains to competing registries, collect fees from the licensees, and in the process establish itself as the manager of the DNS root-all without any formal legal or governmental authorization.

6.4.2 The U.S. Government Claim

The Internet Society's claims did not go uncontested. As soon as the Internet Society began to circulate its IANA charter early in 1995, Robert Aiken, the U.S. Energy Department's representative on the Federal Networking Council (FNC), began to ask uncomfortable questions. In a March 1995 email message that went out to the IETF, the Federal Networking Council, the Coordinating Committee on Intercontinental Research Networks, and the ISOC board, he asked, 'Is ISOC claiming that it has jurisdiction and overall responsibility for the top-level address and name space? If yes, how did ISOC obtain this responsibility; if no, then who does own it?' ^{[50 ]}

In his reply to Aiken, Vint Cerf argued that the Internet was becoming increasingly international and public in character and that management of the name and address space needed to adjust: '[I]t seems to me as if it is possible to make some deliberate agreements now among the interested parties (among which I would include the NICs, the IANA, the various U.S. Gov't research agencies, and ISOC) as to how to proceed in the future. My bias is to try to treat all of this as a global matter and to settle the responsibility on the Internet Society as an nongovernmental agent serving the community.' ^{[51 ]}No formal decision seems to have emerged from these exchanges. They did, however, prompt the National Science Foundation to sponsor a conference on the 'coordination, privatization, and internationalization' of the Internet in November 1995.^[52] The event brought together many of the key participants in Internet administration.

At that conference, Mike St. Johns, the DARPA representative on the Federal Networking Council, set out a description of authority over the name and number spaces that stood in stark contrast to the one being advanced by the Internet Society. The Defense Department, he asserted, owned the name and address spaces. It had delegated 'ownership' of IPv4 addresses to the FNC 'with the understanding that DOD would continue to have first call on the number space if they needed it, but that block and other delegations would be done by the InterNIC in consultation with the IANA' and other agencies. Policy ownership of the DNS root, St. Johns asserted, was transferred to the FNC at roughly the same time as the number space was delegated. St. Johns believed that policy control of the .com, .org, and .net domains remained with the FNC. According to St. Johns, the InterNIC and the IANA were funded by NSF and ARPA, respectively, and therefore those federal agencies 'maintain both fiduciary and program responsibilities' for them. ^{[53 ]}Other comments reveal that both Aiken and St. Johns were critical of the Internet Society and felt that it lacked the ' international standing' to take over authority for the root.

The non-U.S. participants were not pleased. Reacting from a European perspective, Daniel Karrenberg of RÈseaux IP EuropÈens (RIPE) asserted that 'the IANA, not the InterNIC' owns the address space and urged everyone to 'take an international perspective.' ^{[54 ]}David Conrad, representing the newly created regional address registry for the Asia-Pacific region, voiced similar sentiments. Even within the United States, most members of the technical community, particularly Cerf and Postel, were deeply uncomfortable with assertions of national authority over Internet administration.

6.4.3 The Broadening Dialogue

The November 20, 1995, event proved to be the first of a series of conferences and workshops on Internet governance that continued throughout the year 1996. The conferences expanded the dialogue beyond the Internet engineering community to include representatives of trademark holders, legal scholars, and international organizations such as the International Telecommunication Union (ITU), the Organization for Economic Cooperation and Development (OECD), and the World Intellectual Property Organization (WIPO). This included a February 1996 conference on Internet administrative infrastructure sponsored by the Internet Society and the Commercial Internet eXchange (CIX), a June 1996 meeting sponsored by OECD in Dublin, and a September 1996 conference on 'Coordination and Administration of the Internet' sponsored by the Harvard Information Infrastructure Project, the National Science Foundation, CIX, and the Internet Society.

As soon as draft-postel was put forward as a live option, the proposal encountered vocal opposition from a variety of interest groups. Attacks were made not only on the substantive policy it defined but also on the legitimacy of IANA/ISOC to set policy and to collect funds from the authorization of new top-level domains.

One of the most vehement critics of draft-postel was Robert Shaw, an ITU staff member. Shaw charged that IANA lacked the authority to 'tax the root' ^{[55 ]}and ridiculed draft-postel's informal arrangements: 'According to Postel's draft, these potentially multimillion-dollar-generating registries will be awarded by an ‘ad hoc working group' [who are] for the most part engineers [with] no real legal or policy framework behind them' (Shaw 1997). A deeper agenda underlay the ITU's interest in domain name issues. As the intergovernmental organization that had presided for decades over a regime of state-owned telephone monopolies (Cowhey 1990), the ITU was uncertain of its role and status in a new, liberalized order. With the Internet on the rise, private-sector-led standards forums proliferating, and the days of traditional, circuit-switched telephone service seemingly numbered, the ITU needed to assert a role for itself in Internet governance or standards setting. The governance debates presented it with an opportunity to establish itself as an actor in that arena.

Trademark holders also objected to draft-postel's expansion of the name space, although their role was not as prominent at this juncture as it would be later. They feared that it would increase the scope for name speculation and trademark dilution, and that mark holders would feel obliged to register their names in all new domains (Maher 1996). At this time David Maher, co-chair of a new Committee on the Internet formed by the International Trademark Association, emerged as one of the spokesmen for the trademark community on domain name issues. Maher had served as trademark counsel to McDonald's Corporation and in that capacity had facilitated the highly publicized transfer of mcdonalds.com from the journalist Joshua Quittner to the company.

Draft-postel even failed to win the support of the prospective domain name registration businesses, despite its plan to authorize hundreds of new registries. By late October 1996, the alternative registry operators had become completely disenchanted with the IANA-led process and had begun to voice explicit attacks on Postel and the process that had produced the draft. ^{[56 ]}What had begun as complaints about the fees required to enter the market, and IANA's and the Internet Society's authority to assess them, evolved into a deeper challenge to the whole IANA model of DNS administration, with a single, authoritative root zone file set by a central authority. Leading critics such as Karl Denninger argued that rights to top-level domains should be established on a first-use basis by registry operators and that the root servers supporting those registries could be coordinated on a voluntary basis: 'The problem [IANA] people have with this scheme is that it undermines the control structure that some people just don't want to give up. Specifically, if you have a dozen TLD consortia, defined by the root name server 'sets,' then NOBODY-not IANA-not ALTERNIC- not MCSNet-not ITU-not ANYONE-can dictate to people what the fees or market forces are that cause TLDs to exist.' ^{[57 ]}

In a widely read article, a columnist in CommunicationsWeek with ties to the alternative root operators attacked the 'Net governance cartel' and dismissed draft-postel as an 'Amway-style multilevel marketing scheme whereby IANA would essentially franchise TLDs, collecting a piece of the action from downstream distributors while maintaining authoritative control' (Frezza 1996). The newdom list degenerated into a shouting match between supporters and detractors of Postel/IANA. Paul Vixie, writer of the BIND code and a member of the Internet old guard, accused the alternative registries of an attempted coup: 'Rather than work within the process (which would at this point mean attending some ISOC open board meetings) they are attempting a coup. I think IANA's done a fine job for a decade and that it is insulting, to say the least, for folks to try a power grab when the IANA's open/public change process is just about complete. . . . The people who want to pull [the DNS root] away from IANA are not in this for your revolution, man, they're in it for the money.' ^{[58 ]}

By the fall of 1996 it was clear that Postel and the Internet Society lacked the legitimacy and support needed to implement their plan. But no other claimant with wider support emerged. Aware of the strong resistance from international networking entities to a U.S. government claim, the federal government took no action to advance or renege on St. Johns' statements. The Federal Networking Council seemed paralyzed; its advisory committee repeatedly sent it strongly worded messages urging it to transfer policy authority over top-level domain administration from the National Science Foundation to 'some appropriate agency,' but nothing happened. The alternative root server confederations could not get Network Solutions or Postel to add their new top-level domains into the root, and they lacked the broad support required to provoke a coordinated migration to a new root server system.

[1] Draft-postel is unfortunately vague about how it would handle conflicting applications for the same character strings; rather than specifying an auction procedure it implies that IANA would use its own discretion (Postel 1996, 19).

^{[45 ]}Interview with Scott Bradner, July 19, 2000. Bradner blames Rutkowski for many of the tensions, claiming that he acted as if the IETF was a 'wholly owned subsidiary of the Internet Society' during his tenure as director.

^{[46 ]}IAB minutes, December 1994, <ftp://ftp.iab.org/in-notes/IAB/IABmins/IABmins.941209 >.

^{[50 ]}Email, Cerf to Aiken, March 18, 1995, <http://www.wia.org/pub/postelianadraft9.htm>.

^{[51 ]}National Science Foundation and Harvard Information Infrastructure Project, 'Internet Names, Numbers, and Beyond: Issues in the Coordination, Privatization, and Internationalization of the Internet,' November 20, 1995, <http://www.ksg.harvard.edu/iip/GIIconf/nsfmin1.html >.

^[52]M. St. Johns, 'FNC's Role in the DNS Issue,' Internet Numbering Issues, Kennedy School at Annenberg Program Offices, Washington, D.C., November 20, 1995.

^{[53 ]}Minutes of the NSF/Harvard conference, <http://www.ksg.harvard.edu/iip/GIIconf/nsfmin1.html >.

^{[54 ]}G. Lawton, 'New Top-Level Domains Promise Descriptive Names,' Sun-World Online, September 1996.

^{[55 ]}IANA's critics charged that it had moved forward with implementation of draft-postel without obtaining IETF approval of it as an RFC and also that it proposed to create an ad hoc working group appointed by Postel rather than an open, IETF working group.

^{[56 ]}Karl Denninger, newdom post, October 28, 1996.

^{[57 ]}Email, Vixie to newdom, October 28, 1996.

^{[58 ]}'The FNCAC reiterates and underscores the urgency of transferring responsibility for supporting U.S. commercial interests in iTLD administration from the NSF to an appropriate agency.' Draft Minutes of the Federal Networking Council Advisory Committee Meeting, October 1996, <http://www.itrd.gov/fnc/ FNCAC_10_96_minutes.html>. According to Mike Roberts of Educom, 'The motion we passed expressed the strong desire that the FNC work hard NOW to develop a sound future foundation for the domain name system when the NSI agreement ends in less than 18 months, and further, that an ‘appropriate entity' be identified to hold responsibility for those parts of the DNS that are found to require permanent stewardship, i.e., not to be handed over for dissection by the greedy private sector types that lust after the alleged NSI monopoly profits.' Mike Roberts to IETF list, November 10, 1996.

< Day Day Up >