July 8, 1999

The Honorable Thomas J. Bliley, Jr.
Chairman
The House Committee on Commerce
2125 Rayburn House Office Building
Washington, D.C. 20515

Dear Mr. Chairman:

On behalf of the Internet Corporation for Assigned Names and Numbers ("ICANN"), please find enclosed responses to the questions set forth in your letter of June 22, 1999, along with various supporting materials.

The issues raised by your letter are matters of genuine public interest that deserve the attention of this Committee. I hope that your inquiry will serve to educate and inform a broader public than is now aware of the complexities of the privatization process of which ICANN is an integral part. Unfortunately, the subject matter involved, the Domain Name System ("DNS"), is not easy for the general public to understand. As a result, the debates and discussions surrounding the attempt to create a private, non-profit entity to replace the management and funding of the DNS historically provided by the United States Government ("USG") have largely been confined to a small circle of already knowledgeable persons, many with very specific interests. Given the significance of this effort, it deserves more attention by a broader audience.

As described in more detail below, the attempt to create a global private-sector entity to serve as a vehicle for determining consensus across the Internet community, and then to manage the implementation of that consensus, is an extremely difficult task. There are no clear models to follow; global organizations today tend to be either private organizations with no requirement to produce consensus, or intergovernmental organizations formed by treaty or international agreement. ICANN, by contrast, is intended to be a non-governmental body, but also to be a vehicle that would reflect international consensus about the management of an important component of a truly global resource -- the Internet.

Thus, to a significant extent, ICANN is a great experiment, and like all experiments, it will go through a period of trial and error. ICANN is now a little over six months old, and we have already experienced both. Thus, we welcome the opportunity to share our experiences with the Committee and with the American public, in the hope that this discussion will help to make this effort more successful more quickly.

Ultimately, ICANN's value to the Internet community lies both in what it does (that is, substance) and how it does it (process). As to substance, if it is successful in developing and implementing consensus on DNS operational and policy issues, ICANN will help ensure the smooth and stable operation and growth of the Internet by providing a workable mechanism for oversight of a select set of key technical administrative functions: the management of the domain name and root server systems; the allocation of IP addresses; and the coordination of technical standards. In addition, there is a clear consensus in the Internet community favoring the prompt introduction and promotion of competition in the delivery of domain name services, and ICANN is charged by the Internet community with accomplishing that goal as well.

As to process, ICANN represents a new approach: non-governmental, private-sector policy-making that is open and transparent, bottom-up, consensus-based, and global in scope. This is difficult enough, but it is taking place in an environment in which the use and importance of the Internet for both personal and business use is growing exponentially. Under these circumstances, and because by definition this process is intended to move forward only on the basis of a global consensus, it is, to put it simply, a tough job. Still, despite the complexity of the task, there has been significant progress, but there are clearly substantial problems left to deal with; your inquiry provides a very helpful opportunity to describe both the progress and the problems for the benefit of the Committee and the entire Internet community.

In addition to responding to the specific questions in your letter, I thought it would be useful to describe, as a common starting point, the context in which ICANN was formed and in which it operates. There is a considerable amount of misinformation out and about, which may have generated some of the questions that the Committee has asked about our progress.

ICANN is a Consensus-Driven Organization. Most importantly, ICANN is a voluntary, consensus organization; it has no statutory authority, and no power at all that is not derivative of the level of Internet community consensus that its policies and procedures represent. It is governed by its bylaws, which were themselves developed over several months of discussion by the Internet community and represent, as the USG determined in recognizing ICANN as the privatization vehicle it had called upon the community to create, the consensus of that community. Thus, ICANN has no power or authority to impose anything on anyone, and it has not attempted to do so. ICANN is nothing more than a vehicle or forum for the development and implementation of global consensus on various policy issues related to the DNS.

Because there were at the time of ICANN's formation and remain today critics of either its bylaws or particular actions taken since its creation, it is useful to define what we mean when we use the word "consensus." It obviously does not mean "unanimous," nor is it intended to reflect some precise counting of heads pro or con on a particular subject, since in this environment that is simply not possible. What it does mean is that, on any particular issue, proposed policies are generated from public input and published to the world at large, comments are received and publicly discussed, and an attempt is made, from the entirety of that process, to articulate the consensus position as best it can be perceived.

Obviously, to the extent any individual or group undertakes to articulate a consensus of the overall community, its work is useful only to the extent it accurately reflects the consensus. ICANN is no exception to this rule. Unfortunately, there is no litmus test that can objectively render a judgment as to whether this standard has been met in any particular situation. Perhaps the best test is whether the community at large is comfortable with the process and the results, and the best gauge of that is probably the level of continuing participation in the process, and voluntary compliance with the policies produced by that process.

This is, necessarily, a more ambiguous standard than counting votes or some other objectively measurable criteria, and it inevitably means less efficient, more messy, less linear movement, as the perceived community consensus shifts and adapts to change, or as perceptions of that consensus themselves are refined or change. Such a process is easily subject to criticism and attack by those not satisfied with the process or the results; after all, in the absence of some objective determination, it is impossible to definitively refute claims that the consensus has been misread, and loud noise can sometimes be mistaken for broad support for any proposition advanced.

Certainly there are those who do not accept that particular ICANN policies or decisions to date accurately reflect the community consensus, and there are some who are not comfortable with the process that has been employed to determine the community consensus. No doubt reasonable people can differ on both policy and process, and certainly there are many opinions about practically everything on which ICANN has acted. Still, it appears that the process has actually worked remarkably well considering the difficulty of the task, as measured by the fact that most of the global Internet communities continue to participate in this consensus development process.[1]

If ICANN were not reasonably successful as a consensus development vehicle, it would simply disappear; since it relies for its existence on voluntary compliance and cooperation by diverse parties around the world, ICANN cannot survive without broad support throughout the global Internet community. The fact that the privatization process of which ICANN is such an integral part continues to move forward, and that most of the constituent elements of the relevant community appear to support continued progress, is strong evidence that, despite the inherent ambiguity and messiness of the process, it is basically moving in the right direction.

This is certainly not to say that these efforts could not be improved; we are all learning as we go. Everyone involved in this effort has made mistakes, and will probably make more as we move along. The future will hopefully be smoother than the past, but we should all realize that in this effort, as in many other difficult enterprises, the perfect is the enemy of the good.

All those involved in the management of ICANN, from the Directors to the volunteers working in its constituent bodies to its very hard-working staff, remain open to suggestions for improvement. Hopefully, the Committee's efforts and the resulting public attention that will be drawn to this complicated but exciting process will help to identify ways that this difficult job can be done better in the future.

ICANN is the Result of a Comprehensive USG Policy Development Process. In January 1998, the USG issued the "Green Paper," which was a preliminary draft of a plan for transferring management of the domain name system from the USG to the private sector. It recommended that the global Internet community create a United States-based, but globally-representative, non-profit corporation to manage the DNS. The Green Paper also outlined several other proposals, including the creation of new generic top-level domains ("gTLDs") and a competitive system of registries and registrars.

After receiving extensive comments from a wide variety of sources, ranging from individuals to foreign governments, from commercial entities to non-profit organizations, the USG issued the "White Paper," which responded to those comments by eliminating many of the suggestions relating to continued USG involvement that had been part of the Green Paper. The White Paper did, however, continue to urge the private-sector Internet community to form a global, consensus-driven, non-profit corporation to carry out DNS management and related policy functions. The White Paper outlined four guiding principles that the USG would follow, and to which the new corporation should be committed -- stability (to maintain and improve the impressive record of Internet stability), competition (to encourage innovation, consumer choice and satisfaction, and lower costs), private sector, bottom-up coordination (to ensure flexibility and reflect the Internet's bottom-up traditions), and representation (to reflect the functional and geographic diversity of the Internet and its users, and to ensure international participation in decision-making processes).

The White Paper suggested that the new global consensus corporation it envisioned should be structured "to equitably represent the interests of IP number registries, domain name registries, domain name registrars, the technical community, Internet service providers (ISPs), and Internet users (commercial, not-for-profit, and individuals) from around the world." Finally, it suggested that it take several early actions, including: (1) "appoint, on an interim basis, an initial Board of Directors," which would serve "until the Board of Directors is elected and installed;" (2) "establish a system for electing a Board of Directors . . . that insures that the new corporation's Board of Directors reflects the geographical and functional diversity of the Internet, and is sufficiently flexible to permit evolution to reflect changes in the constituency of Internet stakeholders;" (3) "develop policies for the addition of TLDs, and establish the qualifications for domain name registries and domain name registrars within the system;" and (4) "restrict official government representation on the Board of Directors without precluding governments from participating as Internet users or in a non-voting advisory capacity."

The White Paper also committed the USG to take certain steps to "accomplish the objectives" set forth in the White Paper. These included (1) "ramp down the cooperative agreement with NSI with the objective of introducing competition into the domain name space"; (2) "enter into agreement[s] with the new corporation under which it assumes responsibility for management of the domain name space"; (3) ask the World Intellectual Property Organization (WIPO) to "convene an international process. . . to develop a set of recommendations for trademark/domain name dispute resolutions and other issues to be presented to the Interim Board for its consideration as soon as possible"; (4) "consult with the international community, including other interested governments"; and (5) "undertake . . . a review of the root server system to recommend means to increase the security and professional management of the system."

ICANN is a Product of Internet Community Consensus. After the issuance of the White Paper, the Internet community began discussions about the shape and nature of this new not-for-profit corporation that would manage the DNS and related functions. As this process was described in the most recent issue of the Harvard Law Review, "In the bottom-up, consensus-building tradition of the Internet, a broad-based coalition of Internet associations, including NSI [Network Solutions, Inc.] and the ISOC [The Internet Society], initiated a worldwide forum, the International Forum on the White Paper (IFWP), to discuss the various implementation issues left unresolved by the White Paper."[2] The IFWP held a series of worldwide meetings that the Harvard Law Review called "akin to a series of traveling constitutional conventions;"these were held in Reston (Virginia), Geneva, Singapore, and Buenos Aires. At the Geneva meeting, "a consensus emerged that distinguished individuals should govern the new corporation, and interest groups should participate in councils [now called Supporting Organizations] addressing specific issues."

By the end of summer 1998, the various Internet communities had come to a consensus on the structure of the new global, consensus, Internet corporation. In fact, NSI and the Internet Assigned Numbers Authority (IANA), the two USG contractors that had historically been responsible for the DNS management functions that were to be transferred to the new corporation, jointly published for public comment draft bylaws for the global consensus corporation that eventually became ICANN. Those bylaws, with minor changes, ultimately became the basis of the proposal submitted to the USG in October 1998 as the Internet community consensus response to the White Paper's challenge. Following another public notice and comment period and some modifications designed to further increase the transparency of ICANN's operations, the ICANN proposal was accepted by the USG through the signing of a Memorandum of Understanding ("MOU") with ICANN.

Simultaneously, the USG agreed (in what is known as Amendment No. 11 to its Cooperative Agreement with NSI) to extend its contractual relationship with NSI from September 30, 1998, to September 30, 2000. This extension was granted to NSI on the condition that NSI cooperate with what became ICANN, and that it begin development of a system to support new competitive registrars, separate its registry operations from its registrar operations (so as to insure that its registrar business would not have an unfair competitive advantage in interacting with its own monopoly registry business), create a searchable domain-name database, and provide technical assistance to ICANN. This approach -- to extend the existing NSI agreement -- was intended to smooth the transition to ICANN of the USG's management responsibilities. It anticipated further negotiations between NSI and the USG on various terms of the transition to competition, including the price and other terms by which all registrars would have access to the monopoly registries still operated by NSI.

The MOU between the USG and ICANN recognized ICANN as the global, not-for-profit consensus organization that the USG had, in the White Paper, called upon the Internet community to create. It also set forth a process for the anticipated transition from government to private-sector DNS management, and it restated that it expected this transition to be completed no later (and hopefully sooner) than October 1, 2000. It set forth the various areas in which the USG and ICANN would jointly work to accomplish the transition, including most importantly (1) "establishment of policy for and direction of the allocation of IP number blocks"; (2) "oversight of the operation of the authoritative root server system"; (3) "oversight of the policy for determining the circumstances under which new top level domains would be added to the root system"; and (4) "coordination of the assignment of other Internet technical parameters as needed to maintain universal connectivity on the Internet." The MOU, which noted that the parties would "abide by" the four motivating principles set forth in the White Paper, was based on the USG's finding (set forth in the MOU) that ICANN was "the organization that best demonstrated that it can accommodate the broad and diverse interest groups that make up the Internet community."

ICANN is Not a "Regulator." As this history establishes, and its bylaws make clear, ICANN is a creation of the Internet community itself; perhaps the best analogy, although not perfect, is a private standards-setting body. It has no statutory authority, and never will; its influence derives solely from the willingness of the various participants in the Internet -- both governmental and non-governmental -- to participate in the development of its policies and abide by the results of that consensus-development process. The global Internet is a voluntary network of (mostly private) networks, and it works in large part because the participants choose to work together to make it work.

There have been, of course, a variety of governmental participants in various aspects of DNS management. For example, the initial assignment of IP addresses -- the essential building-blocks of the DNS -- was carried out by ICANN's predecessor organization IANA, along with various other technical tasks. Historically, this has been done pursuant to contracts with the USG; now that the IANA staff have been absorbed into the ICANN structure, that responsibility is ICANN's -- but without any funding support from the USG.

The management of the various global top-level domains ("gTLDs") is undertaken by various USG entities or contractors; the most relevant to this discussion is the cooperative agreement between the USG and NSI that allows NSI to operate the registries for .com, .net., .org and .edu. While this function was originally funded by the USG, for several years it has been funded by simply allowing NSI to charge the general public a fee for every domain name registered anywhere in the world; currently, that fee is a minimum of $70 per registration, and new registrations are now being created at a rate of approximately 4 million a year -- and growing rapidly.

Finally, national governments have a wide range of roles in the operation of the country-code top level domains ("ccTLDs"); for example, the .us domain is currently operated by the University of Southern California's Information Sciences Institute ("ISI") (as a subcontractor to NSI) under the direction of the USG. Thus, there are various governmental interests in the DNS, but in general, the DNS functions today as a voluntary cooperative effort of a large number of mostly private entities without significant direct USG funding.

This is why the clear consensus of those filing comments with the USG in its policy development proceedings leading up to the White Paper was that no government should attempt to control or regulate the DNS, but instead a private-sector, voluntary management organization should be constructed. The only realistic alternative to such a private-sector approach, with something like ICANN as the consensus development and implementation vehicle, is much broader governmental involvement than now exists. Given the global character and importance of the Internet, this would almost inevitably take the form of some type of multinational governmental entity.

The process that has produced ICANN, and is now moving forward toward the completion of the ICANN structure and processes, is designed to make that multinational governmental approach unnecessary. Once fully functional, ICANN will operate along the lines of how the Internet has always functioned -- through the voluntary cooperation of large numbers of participants -- with the exception that the very informal arrangements of the past will likely be replaced with a series of consistent contractual relationships with the various relevant DNS entities: name registries and registrars, address registries, root-server operators, standards bodies, and (at least for the immediate future) the USG.

If they come into existence, these contracts will be the product of voluntary agreements; since ICANN has no governmental power, and indeed no existence outside the context of community consensus, it cannot coerce cooperation. If such a series of contracts is created, that will be both evidence of the success of this consensus-development process and a strong incentive for those who wish to benefit from connection to this network of networks to comply with ICANN policies -- which will by definition be nothing more than a reflection of community consensus. Indeed, this is the entire objective of the privatization process: to replace a mixed governmental/private informal system of management with a wholly private, global consensus management system reflected in more formal agreements between ICANN and the various DNS infrastructure and other participants.

Thus, ICANN will never be a "regulator." If it is successful in encouraging and accurately recognizing consensus, it will attract the participation of people and entities that want to see the DNS process continue to function effectively, efficiently, and fairly. The more successful this process is, the more influence ICANN's policies will have. But that is as it should be; the broader the consensus, the more powerful the influence and the more attractive the processes and organizations (like ICANN) that are part of that consensus. Thus, the only "authority" that ICANN will ever have is its attractiveness to members of the Internet community as a device for development, articulation and implementation of community consensus. Any "power" that ICANN ever achieves will flow solely from the fact that it reflects the consensus views of the Internet community -- a highly desirable result, and the sole objective of those currently involved in ICANN.

I apologize for the length of this overview, but I hope it will serve as useful context in your evaluation of the progress to date in this ambitious and complex global privatization initiative. The responses to your specific questions are enclosed. Please let me know if we can provide any additional information.

Sincerely,

Esther Dyson
Interim Chairman of the Board
ICANN

Enclosures

Endnotes:

[1] It would take too much space to list the hundreds of individual members of the Internet community who are actively participating in the ICANN process. However, a very limited and non-exhaustive list of some of the groups and companies that have been constructively participating in the ICANN consensus-formation process demonstrates the breadth of the Internet community's involvement and commitment to make the ICANN process a success: Ascend Communications; AFNIC; AfriNIC; America Online; the American Intellectual Property Law Association (AIPLA); the American Internet Registrants Association; Asia & Pacific Internet Association (APIA); Asia Pacific Network Information Centre (APNIC); Association of European Brand Owners (MARQUES); Council of the Asia Pacific country code Top Level Domains (APTLD); the American Registry for Internet Numbers (ARIN); the Association for Computing Machinery (ACM); the Association of Internet Professionals (AIP); AT&T; Bell Atlantic; the Berkman Center for Internet & Society at Harvard Law School; British Telecommunications; the Center for Democracy and Technology (CDT); the Center for Global Communications (GLOCOM); Centraal Corp.; Cisco Systems; the Commercial Internet Exchange (CIX); Compaq Computer Corp.; Concentric Network Corp.; the Council of European National Top Level Domain Registries (CENTR); Deutsche Telekom; the Domain Name Rights Coalition; Dun & Bradstreet; EDUCAUSE; Electronic Commerce Europe; the Electronic Frontier Foundation; the Latin America and Caribbean Federation for Internet and Electronic Commerce (eCOM-LAC); European ISP Association (EuroISPA); European Telecommunications Standards Institute (ETSI); Foro de Redes de America Latina y el Caribe (ENRED); France Telecom; Fujitsu; Fundacion Airtel; GTE Internetworking; IBM; the Information Technology Association of America (ITAA); the International Chamber of Commerce; the International Trademark Association (INTA); the Internet Council of Registrars (CORE); the Internet Engineering Task Force (IETF); the Internet Society (ISOC); KPN; MCI Worldcom; Microsoft Corporation; the Motion Picture Association of America (MPAA); Netscape Communications Corp.; Novell; Oracle Corporation; the Organization for Economic Co-operation and Development (OECD); PSINet; RIPE; Sun Microsystems; Symantec; UUNET; Verio; World Information Technology and Services Alliance (WITSA); the World Intellectual Property Organization (WIPO); the World Wide Web Consortium (W3C).

[2] Developments - The Law of Cyberspace, 112 Harv. L. Rev. 1574, 1671-72 (1999).

Page Updated 05-Oct-2002
©1999 The Internet Corporation for Assigned Names and Numbers. All rights reserved.