|< Day Day Up >|| |
The new regime's place in the international order is more stable than it was in 1999, but it is not necessarily secure. In subsequent sections, I briefly discuss the issues and forces that can change or alter the new international regime. The discussions are deliberately sketchy because they address issues that are changing rapidly.
The fight over at-large membership, set in motion during the International Forum on the White Paper process and carried forward throughout ICANN's first three years, is one of the most important arenas of potential change. How that division is resolved will do much to define the nature of the new institution. The issue looms large because it is a battle over competing conceptions of ICANN's identity. The concept of an open, at-large membership was the only truly innovative aspect of the White Paper process. If implemented, ICANN would be the only international regime to incorporate such a high level of public participation.
To accept the need for broad-based public representation is to come fully to grips with the fact that the new organization's purpose is not just technical coordination but the development of economic and regulatory policies. It would constitute recognition that the new institution's legitimacy requires popular accountability, not claims of an insider community's consensus. Those are hard concessions for some stakeholders to make. The technical community would have to relinquish its privileged place in the regime, and the power of business stakeholders would be diminished relative to that of users. But the legitimacy of the organization as a whole would be greatly enhanced.
On the other hand, an active at-large membership capable of electing half the board might set in motion longer-term political and organizational dynamics. Johnson and Crawford are right that global membership and elections tend to bring with them assumptions about the delegation to ICANN of quasi-governmental powers. We have almost no experience with the behavior of a global electorate. We do know, however, that ICANN's first invocation of a global public vote during the fall 2000 elections set in motion competition among national blocs within regions.
Unfortunately the Johnson-Crawford theory does not provide much guidance here. There is a direct contradiction between their belief that the ICANN regime is voluntary and lacks sovereignty, and their fear that an Internet public able to elect a significant number of board members will lead to wealth-redistributing legislatures and majority-rule threats to Internet stakeholders' rights and interests. Either ICANN as an institution has the power to do evil, or it does not. If it does, doesn't the broader public have a right to some direct representation in its governance? The public's rights and interests also can be harmed. If it does not, then what is there to fear from public representation? In a voluntary regime, the major stakeholders would simply opt out if things went awry.
The informal way ICANN was created may have helped it to come into being faster and may have given its backers more flexibility. But by skipping a step or two in the formation of the regime, the process created fundamental ambiguities and tensions about ICANN's legal standing and its relationship to the international order, which is still primarily composed of nation-states and their treaty organizations.
In the past, international regulatory regimes have been formed by treaties among governments. One can expect, therefore, that there will be long-term pressures from governments and other parts of society to formalize and clarify ICANN's relationship to governments. Governments, by virtue of their apparatus of representation, consider themselves to be the legitimate representatives of the general public interest. Evidence of a tension between this self-conception and ICANN's role has already surfaced in the GAC's attempt to insert itself into the country code delegation process.
The Commerce Department's reservation of ultimate policy authority over the root is a ticking time bomb that must either be defused carefully or allowed to explode unexpectedly at some point in the future. The political obstacles within the United States to relinquishing policy authority are legion. To do so would require bringing the entire ICANN and DNS can of worms before the U.S. Congress and possibly refighting the domain name wars in that arena. Nationalistic forces within the United States could easily intimidate proponents of yielding control by accusing them of 'giving away the Internet.' Even non-nationalists who have been close observers of ICANN's behavior in its first three years are not enthusiastic supporters of removing it from the last remaining form of public oversight. And yet, the Commerce Department reservation contradicts almost everything ICANN was supposed to be: private, internationalized, selfgoverning.
The relationship to nation-states will be strongly affected by the resolution of the at-large membership issues. If ICANN manages to avoid fulfilling its promise to create an at-large membership with direct input into the board, then it would be easier for politicians dissatisfied with its policies to claim that it is little more than a cartel of the domain name supply industry. And if that happens, it will be difficult for ICANN to resist attempts by governments to seek more direct and formal oversight capabilities over its actions. On the other hand, if ICANN does develop its own organic capabilities to represent the broad public interest, its resistance to interference from governments will increase. But this would be achieved at the price of becoming more governmental itself.
Internationalizing domain names-more precisely, using non-Roman scripts as labels in the DNS-would completely transform the domain name registration industry. Multilingual domain names are extremely popular with the large majority of the world's Internet users who are not native English readers. Implementing a capability to resolve non-Roman scripts could require major changes in the DNS protocol. An IETF working group is over a year behind schedule as it confronts the complexity of human languages and problems with intellectual property rights over certain techniques. [7 ]Some methods of internationalization (known as serverside approaches) would redesign the entire DNS to fully accommodate new scripts, and would probably take a decade to be fully implemented. Less radical changes in DNS, implemented at the client side as software plug-ins into users' browsers, would lead to faster implementation but pose other problems.
As is the case with any change in a technical standard, universal interoperability is at risk as transition and migration take place. Uncertainty in the marketplace can encourage competing vendors supporting different standards to seek market share that may enhance their leverage over the process. Implementation of internationalized domain names could lead to a multiplicity of new roots. Because linguistic differences often come bundled with differing political and economic interests, the implications of internationalizing DNS are not to be taken lightly. [8 ]
Multilingual domain names may also put pressure on ICANN to expand its mission. During ICANN's creation, key members of the IETF hierarchy adamantly insisted that standards development was IETF's turf and ICANN had no place in it. Now, however, Internet Architecture Board members who are fearful of competing internationalized domain name standards have suggested that ICANN exploit its ability to regulate registries and domain names to promote a specific standard. [9 ]In the revision of the Verisign contracts, ICANN's legal counsel, Joe Sims, also made it clear that ICANN would like the contracts to explicitly acknowledge ICANN's authority to impose standards on registries pertaining to internationalized domain names. [10 ]Thus, ICANN seems poised to exploit its position at the root to regulate standards adoption, just as the Federal Communications Commission and other national regulatory authorities leverage their control of the spectrum to impose technical standards on wireless industry players.
A second WIPO process, initiated only a year after the first one concluded, proposes to create sweeping new rights in names and to use the domain name system to enforce them. If the new WIPO proposals are accepted, the domain name system will take another major step toward becoming a highly regulated arena, and the types of rights recognized will expand from trademark to a host of others. This issue is covered in detail in chapter 11.
Country codes top-level domains (ccTLDs) are in some ways the most conservative element of the new regime, mirroring as they do the political geography of the existing international system. In another sense, they are the most disruptive, autonomous, and decentralizing force. Country code registries owe little to ICANN. Although most of their delegations originally came from Jon Postel, the new organization lacks any clear authority to withdraw them. Indeed, because of the tenuous and sometimes wholly artificial aura of 'national sovereignty' surrounding them, ICANN must tread lightly in its treatment of the country codes. The protection and support of national governments serves as an effective check on the power of the U.S. 'public benefit' corporation.
The ccTLDs represent a hole in the regime, just as Network Solutions did during most of 1999. Currently, ICANN has no contractual relationship with the ccTLD registries. Its ability to tax them to support the regime is contested, as noted in chapter 9. This means that over 200 registries, representing about 20 percent of the world's domain name registrations, really do have an almost entirely voluntary relationship with ICANN. They are free to adopt their own naming conventions, to adopt their own dispute resolution policies, to fund ICANN or not, as they see fit.
WIPO is systematically wooing the ccTLD registries, offering to bring them into ICANN's Uniform Dispute Resolution Policy or to help them develop their own policy. ICANN's management is investing heavily in staff resources to repair relations and rope the recalcitrant ccTLD managers into the new regime. Both ICANN and the country code constituency within the Domain Name Supporting Organization (DNSO) are working on the terms and conditions of a model contract. The ccTLD registries have already proposed to withdraw from the DNSO and form a new Supporting Organization with a guaranteed number of board seats. How this relationship is defined will do much to define the character of the regime.
A less immediate issue is the migration of the Internet Protocol to version 6, with its vastly expanded, 128-bit address space. Interestingly, the policies defined by the address registries for allocating and assigning the v6 address space differ little from those currently used for the IPv4 space. [11 ]The Address Supporting Organization plans to allocate IPv6 address blocks hierarchically through its established structure of regional address registries. Assigned users of IPv6 addresses will be considered custodians, not owners, of the addresses. The registries will still use administrative rationing methods and attempt to avoid stockpiling of addresses by requiring demonstrations of need. Most important, IPv6 does not eliminate the most important scarcity affecting the growth of the Internet: the expansion of routing tables, which imposes a need to assign addresses in a way that permits aggregation of routes. The need for route aggregation is now the major constraint on address assignment, and perhaps also the most important constraint on the growth of the Internet (Huston 2001).
The more interesting question about address allocation is, What will happen if IPv6 does not gain acceptance? Pressures to ration more efficiently the extremely scarce IPv4 address space may lead to calls for a greater role for market forces. A market for addresses could introduce some of the pressures for institutional change that have already affected the domain name system. A failure to transition to IPv6 could also pave the way for the entry of more radical technological alternatives that would be completely outside the control of ICANN.
[7 ]The IETF's Internationalized Domain Name (idn) Working Group first met in November 1999. Beginning in August 2000 it began to miss its deadlines. Its work also was affected by the opening of working 'testbeds' by major industry players, notably Verisign. See <http://www.ietf.org/html.charters/idn-charter.html>.
[8 ]A spokesman for China's Internet Network Information Center (CNNIC), reacting to the announcement by the U.S. company Verisign that it would begin testbed registration of Chinese-character domain names under .com, said, 'A company shouldn't be allowed to provide Chinese domain name registration services in China without the approval of the Chinese government.' He added, ' Related Chinese departments have protested to the Internet Corporation for Assigned Names and Numbers (ICANN) that Chinese-character domain names are quite different from the ASCII (English) ones, since they have unique . . . cultural and historic implications. China is seeking to participate in the formulation of the international standard for Chinese character domain names.' China Online, November 3, 2000.
[9 ]John Klensin, IAB chair, urged ICANN to 'start warning the relevant domains' of the harm being caused by the testbeds and accompany the warning with a threat to 'start a redelegation process' of their domain. Cited in 'Status Report of the Internationalized Domain Name Internal Working Group of the ICANN Board of Directors,' June 1, 2001, <http://www.icann.org/committees/idn/status-report-05jul01.htm>.
[10 ]In a January 22, 2001, email to Roger Cochetti initiating negotiations on revision of the Verisign-NSI divestiture agreement, Sims refers to 'revision of the registry agreement to make it clearer that ICANN has the right and power to set technical standards (a current example would be multilingual).' Posted on ICANNWatch site, 'Text of Joe Sims' ‘Willing to Advocate' Email,' May 17, 2001, <http://www.icannwatch.org/article.php?sid=156>.
[11 ]'Provisional IPv6 Assignment and Allocation Policy Document,' version released May 28, 1999, and amended July 14, 1999, <http://www.apnic.net/drafts/ipv6-policy-280599.txt>.
|< Day Day Up >|| |