As the Timeline shows, most of this period was marked by the operation of the Internet as an amalgamation of Department of Defense network and host computer resources, followed by a period of joint DOD and U.S. Government research agency arrangements. It was under those arrangements that all the present DNS administrative were fashioned.

In early 1995 - roughly coincident with the emergence of substantial commercial interest in the Internet - the backbone network administration was transferred to the private-sector by the relevant Federal agencies. At the same time, responsibility for the basic identifier administration was continued indefinitely by DOD and registration of what had long been regarded as U.S. top level domains - .COM, .ORG, .NET, .EDU, etc. - were retained under a NSF cooperative agreement until the end of the contract period in 1998. In 1995, following considerable interaction with the private-sector, and in the face of scores of other country registries worldwide assessing registration fees to recover costs, the NSF authorized nominal fees for registration.

During the 1994 to 1997 timeframe, a combination of experimentation and self-instituted initiatives by various individuals and institutions resulted in diverse attempts to institute new arrangements for DNS registration and operations. This included individually and in groups: the Information Sciences Institute, AlterNIC, Image On-Line, the Internet Society, the International Trademark Association, the International Telecommunication Union, the World Intellectual Property Organization, Iperdome, PGMedia, among others. None of these self-initiatives achieved significant, broad support by Internet industry or users, and some possessed clearly unlawful and highly undesirable and destabilizing attributes.

In July 1997, the Department of Commerce's NTIA, working together with myriad other Federal agencies via an interagency committee, instituted a notice of inquiry, followed in February 1998 with a notice of proposed rulemaking, concerning the transfer of these remaining Internet identifier administrative responsibilities to the private-sector. Congress also played an important role in conducting public hearings on DNS developments. This activity has now culminated in the release on 4 June 1998 of its Policy Statement on DNS.

The most important part of this transition has always been the transfer of responsibility for all Internet identifiers - rather than just domain names which although they have received considerable frequently artificially generated publicity, are but one of many kinds of names, addresses, and numbers associated with Internet operations. Indeed, the management of Internet Protocol addresses, for example, is more important and contentious, as they are required to actually make the Internet work, and can have significant competitive implications among Internet Service Providers. This has come to the fore, for example, in reviews by the Department of Justice and the European Union in the Worldcom-MCI merger.

.Because of the importance and de facto power associated with this Internet identifier administrative function, it was not only a responsibility retained by the DOD, but also one which was highly prized and sought by various private parties and intergovernmental organizations over the past four years. It was also a highly controversial one where no due process or public decision making records existed, where all the power resided in one individuals with what became minimal Federal oversight, and where technical academic community factions created a religious cause against the perceived encroachment of commercial interest and public interests on "their" Internet.

Coupled with this, were U.N. agencies like the International Telecommunication Union (ITU) and the World Intellectual Property Organization (WIPO) who saw the opportunity to acquire jurisdiction over Internet administrative functions by joining with various factions, declaring Internet identifiers to be spectrum-like global public resources, denigrating the U.S., Federal agencies and contractors, and crafting their own international agreements in the form of a Generic Top Level Domain Memorandum of Understanding (gTLD-MoU) that bypassed all normal legislative processes.

All of this was exacerbated by an Internet industry which was organizationally immature and which traditionally eschewed government bodies and processes. Lastly, impending Federal and state judicial intervention threatened to further complicate this strange mix.

Getting the transition model "right," in the face of this "hot potato" as the New York Times still refers to this subject as of Monday, became a major challenge, and one which was collectively accepted by participants drawn from many Federal Agencies under the combined leadership of Presidential adviser Ira Magaziner and the Department of Commerce NTIA's Larry Irving and his staff. They quickly recognized the problems and issues, crafted a public inquiry and rulemaking proceeding, and devoted enormous energy to work with all the diverse Internet factions to effect the right model.

The NTIA Policy Statement calls for industry to create an open, broad-based corporation and quasi-public policy making and adjudication processes to undertake these responsibilities. This is an entirely new kind of creature for the Internet arena, and the establishment of this model - as well as its success - is an important test of the ability for a highly complex, diverse industry to effect self-governance and deal with difficult legal and policy matters. The implications of this action go far beyond just Internet identifiers - affecting other arenas such as electronic commerce and expression where governmental solutions remain a potential alternative if self-governance doesn't work.

The Statement - in eschewing radio spectrum like models and keeping inter-governmental organizations at bay - has provided a playing field where an industry model can proceed. At the same time, the Statement provides for continuing necessary stability and consumer trust by relying on legacy registry providers under a more competitive model. Network Solutions, Inc., has already made considerable progress along these lines - as outlined at earlier Science Subcommittee hearings,

Industry has responded immediately to the Statement by organizing a two-day open global workshop of all stakeholders and interested parties here in Washington, to develop the model, structure, balance of parties, and provisions for incorporation. People and organizations from multiple regions internationally are being contacted to attend. Fortuitously, the workshop is being led by one of the world's most eminent corporate and fiduciary experts, Prof. Tamar Frankel at Boston University Law School, who has recently been an advisor to the Securities and Exchange Commission. Prof. Frankel has not only published the leading treatise in the field, but worked extensively in other countries in the creation of fiduciary corporations and trusts, and her work has been cited in the highest judicial proceedings of multiple countries.

The challenge for Prof. Frankel, the new corporation, and the industry will be to get all the stakeholders and parties to work together effectively. Annex B contains an outline of all these players. The diversity of perspectives, experience, and objectives is substantial. Hopefully all the stakeholders can set their individual interests and views aside and cooperate in creating and maintaining an open, neutral, balanced, industry-driven forum. This will be required not only at the outset, but on a continuing basis.

As noted above, some staff and participants in the International Telecommunication Union as an intergovernmental organization are aggressively pursuing significant DNS-related jurisdiction and functions through an ITU Memorandum of Understanding on this subject. At the ITU's governing Council meeting just held 20-29 May, this action was apparently terminated through the intervention by the U.S. representative. However, at least one representative from another government promised this matter would be revisited at the ITU's Plenipotentiary Conference of all its governmental members beginning 12 October 1998. Thus the potential exists for the NTIA Statement to be undercut through independent actions taken by government initiative in multilateral bodies, and the U.S. Department of State working together with industry, needs to work together and expend resources over the coming months to assure this does not occur.

Introduction of Competition

The introduction of competition in the domain name system has always been a desirable goal rather than a compelling need. Internet industry and users have always asserted trust, stability, performance, and reasonable fees as more significant, and this is amply reflected in the comments files in response to the NTIA proceeding and in previous Congressional testimony.

Until rather recently, the so-called generic top level domains have always been regarded as de fact United States domains, and there actually more than 200 other top level domains that are generally associated with different countries throughout the world. In the case of several small countries, the authorities have actually leased their top level domains to commercial concerns who market them as alternative domains. Each of these has its own management arrangements, and virtually none of them operate competitively. Indeed, ironically, it is these other national domain registration arrangements that have experienced far greater problems of poor service and high prices than the domains .com, .org, etc.

Over the past two years, the various attempts to foster additional competition through diverse models and techniques by entrepreneurs and consortia such as AlterNIC, eDNS, Image On-line, Iperdome, CORE, AH-Net, Network Solutions Inc, ORSC, have demonstrated that some measure of competition may promote good service and greater choice at reasonable prices to end users. Although the objectives and perspectives among these parties may differ, they have all contributed to advancing alternative new DNS services. There is no reason all of them cannot coexist

However, the creation of additional new TLDs must be weighed against the exacerbation of trademark monitoring burdens and disputes faced by trademark holders, the public interest associated with new more complex names, consumer fraud, and overall system performance.

Although it is often claimed that vast demand exists for new domain names, it is worth noting that the use of leased national TLDs by several parties thusfar has not produced a significant market response. As noted below, users and consumers seems comfortable with the existing arrangements. Studies being contemplated by Congress and others in this regard will be very helpful in understanding the tradeoffs.

Use by Consumers and Business

This dynamic occurs through the ability of users to construct any desired simple language expression and acronyms - to create an identity or make a statement and associate it with the familiar .com, .org, or other generic or country symbols. Even with millions of domains registered in the .COM domain, the enormous flexibility of language generally allows sufficient variants to meet new needs. For example, the industry effort for the new corporation workshop took less than five minutes to construct and register GIAW.ORG

The stability and simplicity of the existing DNS arrangements seem to have met with widespread satisfaction, and few have advocated significant change. Nonetheless, change is inevitable, and the new corporation or one of its subcommittees can serve an invaluable means of achieving a consensus among all the competing parties and interests on desirable directions and provider requirements.

Future Technology and Applications

Already, new object models and mobile agent technologies have produced specifications for next generation domain name systems that will see maturation and implementation over the coming years. World Wide Web, agent, and object standards organizations and developers have proceeded independently in these directions. Some of these new specifications are much more flexible than those of the current domain name system, and need to accommodate self-defining ontologies (knowledge structures) as well as an Internet environment that will consist of countless, ever evolving places and mobile objects.

. At the same time, new directory services are emerging and will reduce the reliance on the existing DNS to locate business, institutions, products, and services. Already, most users today probably find sites through popular search engine sites rather than looking up domain names.

It is important for the Internet administrative arrangements to remain flexible to allow continuing, rapid, self-organizing experimentation and evolution of technology and applications.