U.S. DEPARTMENT OF COMMERCE
National Telecommunications and Information Administration
Washington, DC 20230
In The Matter of )
IMPROVEMENT OF TECHNICAL )
MANAGEMENT OF INTERNET ) Docket No. 980212036-8036-01
NAMES AND ADDRESSES )
Hugh Latimer )
Wiley, Rein & Fielding )
1776 K St., N.W. )
Washington, DC 20006 )
COMMENTS SUBMITTED IN OPPOSITION TO ADDING UP TO FIVE
NEW GENERIC TOP LEVEL DOMAINS
In the past few years, domain names have come to be viewed as source-identifying symbols rather than mere addresses. In fact, they have become the cyberspace trademark. As expressed in the International Trademark Association Internet Domain White Paper: Thus, from the standpoint of the consumer of goods and services, an Internet domain name identifies the company, school, institution or organization to which a seller is attached. In short, domain names, like trademarks, often are used to identify a source of goods or services.
This is particularly true of the .com generic top level domain (gTLD) which has been assigned on a first come, first served basis by a single registry, Network Solutions, Inc. (NSI), under contract with the National Science Foundation. The .com gTLD has become so popular that Network Solutions has registered well over 1 million .com domains in the past five years. Numerous companies have registered the company name, company products, generic terms relevant to the companys business (and undoubtedly many other things) in the .com domain. Such vanity domain names, like vanity telephone numbers, have been viewed as helpful marketing tools since it is thought that consumers are better able to remember and hence locate the companys web site.
As the Internet has grown exponentially, an increasing number of trademark infringement and dilution lawsuits have been filed in the U. S. and elsewhere, including both cyber-pirating of well known trademarks as domain names and more legitimate disputes between parties who desire to use the same or similar trademarks as domain names. Further, NSI itself has been the object of extensive criticism based on dissatisfaction with the way it has handled such domain name conflicts, and has been named as a defendant in a number of cases.
A prime problem with the Green Paper as it relates to the trademark dilemma caused by gTLDs is that it seriously understates the problem. The statement that trademark/domain name disputes arise very rarely on the Internet today, falls far short of reality. In fact, many companies with valuable trademarks will attest that the job of policing and protecting their trademarks from infringement or dilution on the Internet, even as it currently exists, has become pervasive. The problem for trademark owners cannot be measured solely by the number of disputes reported by NSI, since NSI only declines to register the exact same domain name. Similar domain names present equally egregious infringement and dilution problems for trademark owners, and many companies have opted to resolve their trademark protection problems by negotiation (often paying a substantial penalty to stop infringement and dilution), reserving only the most intractable cases for litigation.
The Green Paper proposal that a study be undertaken on the effects of adding new gTLDs and related dispute resolution procedures on trademark owners is an excellent one (as is Senator Leahys recently introduced bill to fund such a study). However, a study, including a comprehensive survey of companies with valuable trademarks, focusing on the trademark policing and protection problems which already have been caused by the .com gTLD, would be more than sufficient to demonstrate the inadvisability of adding any new gTLDs. If more information is thought to be necessary, the addition of a single new gTLD would suffice. Why add five more gTLDs to conduct a study when even the addition of one new gTLD would worsen an already bad situation? An additional problem is that once a gTLD is added and companies rush to register their trademarks in the new domain (with the attendant advertising and promotion of the new domain names thereafter), it would be difficult to phase out, even if the study found that the new gTLD had unduly exacerbated the trademark problems.
Some concern has been expressed about the antitrust ramifications of adding only a single new gTLD (and one new registry). For a number of reasons, such concerns appear to be misplaced. In analogous situations, the courts and the antitrust agencies have declined to find per se antitrust violations where seemingly restrictive arrangements provided procompetitive benefits or cost savings. Thus, in the copyright licensing area, the Supreme Court held that while ASCAP and BMI were literally price setting organizations, their blanket copyright licenses were not naked restraints of trade with no purpose except stifling of competition, but rather their blanket licenses promoted integration of sales, monitoring and enforcement against unauthorized copyright use. The alternative was individual licenses negotiated among thousands of users and thousands of copyright owners. Broadcast Music, Inc. v. Columbia Broadcasting, 441 U.S. 1, 20 (1979). Similarly, in the Domain Name System, as the Green Paper acknowledges, having more than one or two registries would be far more complex technically . . . and it would be much more difficult for trademark holders to protect their trademarks if they had to police a large number of top-level domains. A single registry, or one additional registry, would not result in stifling of competition, since competition would be provided at the next level --the registrar level where there would be an unlimited number of registrars if the Green Paper recommendation is adopted.
This situation also is analogous to production joint ventures between competitors where competition is provided at the next level. The production joint venture between General Motors and Toyota, then the worlds two largest automobile manufacturers, was permitted by the FTC since the joint venture car was less costly to produce and competition was provided at the next level where General Motors and Toyota competed against each other in the sale of the joint venture car. 103 F.T.C. 374 (1984).
Finally, the marketplace itself has demonstrated that it prefers a single registry. With more than 200 national or country code TLDs, customers have selected .com as the domain name of choice. As the Supreme Court pointed out in Broadcast Music, ASCAP and the blanket license developed together out of the practical situation in the marketplace . . . . Id. at 20. Competition among an unlimited number of registrars, vying to provide one or two gTLDs to customers, should alleviate any antitrust concerns.
A long term answer to the antitrust question and, even more, to the trademark dilemma, is to do away entirely with domain names. Vanity domain names have created a new layer of trademarks, up for grabs to the first comer, which has led to numerous disputes between domain holders and trademark owners and to mammoth policing problems for the latter. Such disputes and protection problems need never arise if a numbers-only system is adopted for domain names. Thus, numbers-only domains, like telephone numbers, would come to serve their only essential purpose -- to enable consumers to locate web sites and to receive and send e-mail. This will require development of a worldwide telephone book database on the Internet to provide swift and easy access to everyones domain number.
A numbers-only domain system has been discussed at various times, but summarily rejected with little analysis. In its White Paper, INTA posited that what appears to be a continued interest by the commercial community in vanity domains for marketing purposes, makes it unlikely that a pure random number or alphanumeric system would be implemented. In its response to the Department of Commerces July 15, 1997 Request for Comments, the interim Policy Oversight Committee conceded that it would be impossible to construct a domain name system that is free of trademark conflicts other than meaningless strings of numbers and letters. However, like INTA, it rejected such a system since [t]here appears to be no significant support for having meaningless strings aside from a few proponents who do not take account of the practical and historical support for a mnemonic based system.
The assumed lack of support for a numbers-only system may be the result of not asking the right question. If the choice were posed as between (1) a numbers-only system (with ready access to a worldwide telephone book database) or (2) the addition of numerous new gTLDs (requiring registration in and policing of all such domains), it is not at all clear that support for the former would be insubstantial. Vanity telephone numbers may be helpful marketing tools but most firms and individuals get along without them. Similarly, easy to remember domain names may be helpful marketing tools on the Internet, but unnecessary if an easily accessible telephone book database were available. Ideally, such a database would enable an Internet user to type in the desired individual name or firm and the applicable e-mail number or domain number would appear on the screen. The technology for such a database appears to be available, or will be available in the near future. As the Green Paper points out, in the future, directory services of various sorts will make it easy for users to find the sites they seek regardless of the number of top-level domains. A better solution is to develop the directory services and eliminate the top-level domains.
In sum, adding up to five new gTLDs during the transition to private management of the Domain Name System would only serve to exacerbate the severe problems currently faced by owners of valuable trademarks resulting from the use of vanity domain names on the Internet. The proposed study should focus on the trademark policing and protection problems which already have been caused by the .com gTLD, adding, at most, a single new gTLD to provide more information. If, as expected, the study demonstrates that the problems caused by vanity domain names outweigh the benefits of such a system by a wide margin, serious consideration should be given to phasing out gTLDs and adopting a numbers-only domain system.
March 23, 1998