browse comments: WIPO RFC-3

WIPO RFC-3
mueller@syr.edu
Sat, 6 Mar 1999 00:16:19 -0500

Browse by: [ date ][ subject ][ author ]
Next message: Italian Naming Authority: "WIPO RFC-3"
Previous message: ATT: "WIPO RFC-3"

Attachment: http://arbiter.wipo.int/processes/process1/rfc/dns_attachments/rfc3/attach920697379.doc

Commentary of Professor Milton Mueller, Syracuse University School of Information Studies, on WIPO interim report (RFC-3)

The White Paper asked WIPO to "initiate a balanced and transparent" investigation of domain name - trademark disputes. I commend WIPO for the transparency of its proceedings. The White Paper also asked WIPO for "findings and recommendations." Here I offer no praise. The results of your process are deeply disturbing.

As a set of "findings," RFC-3 is barren of serious factual analysis, and what little evidence it does present is biased.

As a set of "recommendations," RFC-3 takes one's breath away. The proposals are a radical attempt to create an entirely new, unaccountable, global system of administrative law for intellectual property holders. RFC-3 would bring about a massive expansion of the scope and strength of trademark protection. And it exhibits an appalling indifference to basic concerns about fairness to individual users.

The findings
Findings of fact are, or ought to be, a critical part of RFC-3. WIPO proposes sweeping institutional changes and a dramatic realignment of rights. Such changes can be justified only by correctly diagnosing the problem and carefully documenting its scope and scale.

It is disturbing that one member of your expert panel, unaided by WIPO's substantial institutional resources, did a more thorough job of summarizing the types of conflicts, and calmly discussing what is known and not known about their dimensions, than the entirety of RFC-3. I am referring to Professor Michael Froomkin's critique, specifically, paragraphs 42 to 63, subtitled "Types of Conflicts." I propose that that material be incorporated into your final report, verbatim.

Based on that analysis, on publicly available data, and on my own research, I can say with confidence that your estimation of the problem is inflated and your analysis unbalanced. Here are some of the things we know about the problem of domain name-trademark disputes, and which ought to find their way into your report:

* We know that in the open gTLDs, DN/TM conflicts are a very tiny proportion of overall registrations (0.0016)

* The only objective, consistent, longitudinal source of statistical data available-the number of times the NSI dispute resolution policy has been invoked in the open gTLDs-shows that the absolute number of conflicts is going down. It dropped from a peak of 905 in 1997 to 838 in 1998. Because the number of registrations doubled during that period, the ratio of conflicts per 1000 registrations is plummeting (from 1 case in every 1063 registrations in 1997 to 1 case in every 2267 registrations in 1998).

* We know that major TM holders are irritated by name speculators and sometimes incur significant costs fighting them. But evidence suggests that most of those problems are produced by a handful of people. We know that TM owners have won court judgments in nearly every case when name speculators have been challenged.

* We know that Internet-based commerce is one of the fastest-growing parts of the world economy, and that investors, especially individual investors who are representative of the consuming public, are so confident of Internet commerce's potential that they have pushed share prices of Internet companies to extraordinary levels. Thus, there is no factual basis for the partisan claims of trademark holders that domain name misuse is a major threat to the viability of commerce on the Internet.

* We know that fraudulent and deceptive use of domain names of the sort that actually harms consumers (as opposed to irritating TM holders) are much rarer than cases of name speculation. Moreover, what cases exist have been identified and shut down quickly. My own study of publicly reported cases of DN/TM conflict showed that only 12% of them could be classified as infringement, and all incidents had been stopped quickly by settlement or court action. There is reason to believe that the method used in this study would over-represent cases of infringement, because of the greater likelihood of their being documented than other types of conflicts.

* While blatant cases of infringement and fraud do exist, no one has provided a compelling reason to believe that such practices are more prevalent or dangerous on the Internet than they are in any other medium. Because large trademark holders undoubtedly have access to data about the number of violations in other media, their inability to produce comparative data raises doubts about their claim that infringement problems on the Internet are out of the ordinary.

* We know that claims of WIPO commenters such as Bell Atlantic and AT&T that the problem is "increasing regularly" are not substantiated in any systematic or verifiable way. They refer repeatedly to the same anecdotes-individual cases such as "attt.com," all of which have been corrected for months-but provide no continuing stream of data about the number of new cases that could prove an upward trend. Since these companies undoubtedly have access to data about the number of these cases, their inability to produce such data raises serious doubts about the veracity of their claims.

* We know that commercial name protection and registrations have developed that offer multinational trademark owners pre-emptive registrations in all top level domains. RFC-3 does not mention such services or assess their ability to achieve protection without the onerous and compulsory bureaucracy proposed by WIPO.

* We know that a substantial part of the domain name trademark disputes can not and should not be resolved by trademark law at all, but involve "character string" conflicts between legitimate claimants to the name. (Examples: prince.com, veronica.org, pike.com) My study of 121 publicly available cases of domain name trademark conflicts found that 58 cases, or 49%, could be classified as string conflicts. RFC-3 attempted to dismiss these findings by burying them in a footnote and claiming that they are not based on a representative sample that would support "scientific" extrapolations to the entire population of domain names. The issue of statistical perfection can be dismissed as a diversionary tactic. It is currently impossible to draw a truly representative sample of domain name trademark conflicts, due to a variety of practical constraints on the availability of data. The data used by the study is the only type of data available. At any rate, 58 out of 121 cases cannot be explained away as a sampling erro
r. It is interesting to note that RFC-3 did not apply such a rigorous standard, or any standard at all, to evidence submitted by corporate trademark lobbyists, but accepted and reproduced their claims uncritically. The application of a double standard raises serious doubts about the objectivity and fairness of WIPO. I look forward to a correction of this bias in the final report.

* Finally, we know that domain name abuse works both ways: there is substantial evidence that TM holders intimidate or sue perfectly legitimate name holders in order to take away a desirable name. There is a willful avoidance of this fact in RFC-3. The final report must explicitly recognize the existence of domain name hijacking by trademark holders and modify its proposals to protect individual users against such abuse.

All in all, everything we know about domain-name-trademark conflicts suggest that the sweeping changes in Internet administration proposed by RFC-3 are not justified by the nature or scope of the problem. There is no attempt here to assert that a problem does not exist. It is simply that the problem does not justify the measures WIPO proposes. Simple prepayment for registrations and elimination of fraudulent registration information would address the bulk of the problems. The rest can be taken care of in traditional ways. There is no need for a mandatory submission to jurisdiction or any other attempt to create a new system of administrative law.

The recommendations
Paragraphs 32 and 33 of your RFC affirm the principle of parity between TM rights in cyberspace and the physical world. You claim that it is not your goal to create new rights nor to adversely affect other rights such as freedom of expression.

That disclaimer turns out to be meaningless; the recommendations flaunt the parity principle repeatedly. Indeed, I suspect that the RFC protests too much and that WIPO is fully aware of the novelty of what it is proposing. Under the guise of maintaining traditional TM protections in the new venue of cyberspace, RFC-3 proposes to establish uniform procedures and a common set of decision-making principles where there is no uniformity in law. By definition, this creates new rights and a radical disjunction between trademark protection in cyberspace and the physical world.

That disjunction would not be so objectionable if the proposed system were not so egregiously tilted in favor of trademark owners at the expense of domain name registrants. Here is a list of six distinct ways in which the strength and scope of TM protection would be significantly expanded by RFC-3 and rights of free expression diminished:

* RFC-3's TM protection procedures allow trademark owners who are legally entitled to protection that is national in scope and limited to a specific industry sector to claim and win protection and exclusivity that is global and trans-sectoral in scope

* RFC-3 makes access to an essential communications facility contingent upon a statement of purpose and a legally binding promise that registration or use will not infringe or interfere with any intellectual property right. These promises must be made to a globally centralized, exclusive administrative apparatus that can cut off one's access to the Internet with a keystroke. Moreover, this administrative apparatus is capable of instant surveillance of every registration in the world, holds authoritative contact information, and gives any third party trademark holder in the world a right to drag the registrant into a mandatory dispute resolution process. Such conditions and circumstances are not attached to the use of any other medium. We do not have to announce a purpose or make any promises to a global, central registry before leasing a telephone line, purchasing a computer, taking home an audio or videotape, or using a laser printer. Yet all these devices can be and are used to violate intellectual propert
y rights. I see in the procedures of RFC-3 a significant expansion of the strength of trademark protection, and a significant diminution of the right of free expression on the Internet.

* RFC-3 repeatedly asserts that domain name "registration or use" can violate a trademark. Normally, one must know how a mark or name is used before it can be judged infringement. The idea that mere registration of a name, without more, can violate a trademark constitutes a significant expansion of the scope of protection.

* In paragraph 244 WIPO implies that holding a trademark provides an affirmative right to a corresponding domain name. This "right" does not exist in any law that I know of, nor have I ever seen a court decision that articulates and recognizes such a right.

* RFC-3 proposes to reserve, prior to any use or attempt to use, entire blocks of names for well-known marks, and it seeks to set up itself as the arbiter of what is a "famous" or "well-known" mark. No international law or agreement is able to define or recognize what is a "famous" or "well-known" mark; clearly, RFC-3 is creating forms of protection that are new, and stronger than existing trademark rights.

* WIPO defines as abusive and actionable domain name registrations that are similar to a TM in order to attract traffic or attention. But unless commercial use is involved, such registrations may be protected forms of personal, political or artistic expression. Here again, RFC-3's protections go well beyond traditional TM protections, and impinge upon freedom of speech.

In general, the recommendations are designed with only one objective in mind: To make it as easy and inexpensive as possible for a multinational trademark holder to take domain names away from individuals. Registrants will have to risk thousands of dollars to defend their domain names against trademark-motivated challenges. Those who challenge registrations, especially corporations with substantial legal and financial resources at their disposal, risk little by making the challenge and will be able to continue their challenge in national courts if they lose. Individual registrants who have their names taken away by the ADR process, on the other hand, will have no legal standing to challenge the outcome, due to the binding contract they must sign to register a name. Is this what WIPO means by "balance?"

RFC-3 is a travesty of justice. In order to save a handful of multinational trademark holders some litigation costs, it diminishes the rights of hundreds of millions of people, and radically undermines the character of the Internet as a free and open space for innovation. This "cure" is far worse than the disease.

 -- Posted automatically from Process Web site