Introduction
It has been two years since the Uniform Domain Name Dispute Resolution Policy (UDRP) and the Rules for Uniform Domain Name Dispute Resolution Policy (Rules) came into force on 1 December 1999. Perhaps now is a proper time to examine their effects. In fact, the UDRP (hereinafter including its procedural Rules) has been continuously and generally evaluated during the past two years, which to a large extent implies its significance and influence in resolving domain name disputes.
While some people highly praised the UDRP for its flexibility, lower costs and efficiency, some other people criticized it for such demerits as restrictive scope, ambiguous language, biased decision, forum shopping, no cross-examination, lack of appellate procedure, limited remedies and revocability by judgment. It is correct that all these problems with the UDRP are considerable. However, this short essay has no any intention to cover each issue in connection with the UDRP, but rather will concentrate on the question whether the UDRP has provided clear and consistent solutions to domain name disputes through the Providers’ decisions made under it.
This essay will first set about an overall analysis of the UDRP, then enter case study. Based on all of these, the conclusion will be made that: (i) the UDRP has failed to provide a clear solution to domain name disputes; and (ii) the decisions under it are consistent in the macro-sense, and inconsistent in the micro-sense.
Domain name disputes versus the UDRP
Overview ---- a literal and statistic analysis
To begin with, take a look at substantial paragraphs in the UDRP. As to an applicable dispute, a complainant must assert and try to prove that: (i) the disputed domain name is identical or confusingly similar to its trademark or service mark; and (ii) the domain name holder has no rights or legitimate interests in the domain name; and (iii) the domain name was registered and is being used in bad faith. Considering that, in practice, it is much easier for a panel to assess the identity or confusing similarity than rights or legitimate interests or bad faith, the UDRP sets out several more detailed guidelines as well:
First, a domain name holder can evidence its rights or legitimate interests in the disputed name by demonstrating that: (i) prior to any notice of the dispute, it has been using, or has prepared to use, the domain name or the other name corresponding to that domain name with a bona fide offering of goods or services; or (ii) it has been commonly known by the domain name, even without trademark or service mark rights; or (iii) it is making a legitimate noncommercial or fair use of the domain name, without intent for commercial gain to misleadingly divert consumers or to tarnish the trademark or service mark at issue; or (iv) its rights or legitimate rights could been justified under some other circumstances.
Second, on the other hand, a complainant in order to establish the holder’s bad faith may furnish evidence to prove one of these: (i) the holder has registered the domain name primarily for the purpose of selling, renting or otherwise transferring that name to the complainant or its competitors, for valuable consideration in excess of documented out-of-pocket costs directly related to that name; or (ii) the holder has registered the domain name in order to prevent the complainant from reflecting its mark in a corresponding domain name, if the holder has engaged in a pattern of such conduct; or (iii) the holder has registered the domain name primarily for disrupting the business of the complainant as its competitor; or (iv) the holder has been using the domain name with intention to attract, for commercial gain, Internet users to other web sites than the complaint’s, by causing a likelihood of confusion with the complainant's mark in relation to the source, sponsorship, affiliation or endorsement of the holder’s web site or location or of a product or service thereon; or (v) whatsoever can otherwise reflect the holder’s bad faith in registration and use.
From this glance, one can acquire some direct impressions on the UDRP: (i) the UDRP expressly deals with disputes between domain names and trademarks or service marks and it remains unclear whether disputes between domain names and other names (such as personal names) could fall within the UDRP; (ii) the “identical or confusingly similar” element seems clear enough for application and needs no further interpretation; (iii) as to the “rights or legitimate interests” and “bad faith” elements, despite the emphatic specification in the UDRP, it still appears less clearer and so will leave large room for the panel’s interpretation, which is a latent factor that will inevitably undermine the consistency of decisions under the UDRP.
Interestingly, the statistics have from another angle demonstrated the “consistency” of decisions under the UDRP. As of 29 November 2001, in total 3,931 disposed proceedings by the providers, 3,133 (79.7%) decisions had favored complainants (i.e. trademark or service mark owners) instead of respondents (i.e. domain names holders). In total 6,934 domain names involved, 5,425 (78.2%) names were transferred or canceled. By contrast, the statistics carried out at about the same time in last year showed that complainants had won 780 (80.3%) cases in total 971 proceedings by decisions. Therefore, the ratio of the decisions in favor of complainants (or holders) to the total decisions has been keeping at the same level during the past two years. This illustrates that, in macro-sense, the decisions under the UDRP are consistent.
However, our greater concern is whether there existed contradictory outcomes under the UDRP for those cases in which similar facts could be found. To find this, we have to make a positive analysis of cases.
Other names other than trademark names
As described above, the UDRP only focuses on disputes between domain names and trademarks (hereinafter including service marks). The disputes between domain names and other names such as personal names or geographical names are admissible under the UDRP only if those specific names are meanwhile entitled to registered trademarks or common law trademarks. While a panel usually has no choice but to accept a specific name as a registered trademark provided that name has been legally registered and authorized ahead of time, it owns large discretion on whether to admit a specific name to be a common law trademark or not. In a large number of cases involved, the panels preferred granting common law trademark rights in names. For example, in the case of Cedar Trade Associates, Inc. v Gregg Ricks, although the complainant did not enjoyed any registered trademark rights in the name “BuyPC”, the panel affirmed its entitlements in terms of common law and decided that case in favor of it.
But in a few cases, the panels were reluctant to admit the common law trademark rights in names. A prime instance is the case of Manchester Airport PLC and Club Club Limited. In this case, although the complainant had been operating the UK’s third largest airport and had traded as Manchester Airport since 1986, the panel still confusingly asserted that the complainant had failed to allege that it owned a common law trademark and thereby excluded the UDRP from application. Although the cases like this one only took a very small part in the whole, their existence per se has undoubtedly authenticated the inconsistency of different decisions, and further indicated the uncertainty of the answer to whether disputes between domain names and other names could fall under the UDRP.
Identity or confusing similarity
It has been mentioned before that under the UDRP the “identical or confusingly similar” element should cause less trouble than other elements. Insofar as available cases, the respondents scarcely contested the complainants’ allegation that the domain names were identical or confusingly similar to the complainants’ trademarks. Notwithstanding, a little number of cases could be found involving the issue of similarity and the conclusions therein have been contradictory. Here are a couple of examples.
One is the case of Wal-Mart Stores, Inc. v Walsucks and Walmarket Puerto Rico. The complainant owned the “Wal-Mart” trademark that had been well-recognized in the United States, Canada and many other countries, whereas the respondent held several domain names such as <walmartcanadasucks.com>, <wal-martcanadasucks.com>, <walmartuksucks.com> and <walmartpuertoricosucks.com>. In the tedious decision, the panel said that:
“Respondent has appended the term ‘-sucks’ to domain names that are, in the absence of that term, confusingly similar to Complainant’s mark. The addition of the pejorative verb ‘sucks’ is tantamount to creating the phrase ‘Wal-Mart Canada sucks’ (and comparable phrases with Respondent’s other ‘-sucks’ formative domain names)… The addition of a common or generic term following a trademark does not create a new or different mark in which Respondent has rights.”
After browsing among several prior cases, the panel confirmed the existence of similarity, regardless of the respondent’s persuasive contention that a reasonable Internet user viewing a <-sucks.com> formative domain name would assume that the purported trademark owner was not the sponsor of or associated with a web site identified by such a domain name and as a result the <-sucks.com> formative domain names could not be confusingly similar to that trademark.
As opposed to the above one is the case of Compusa Management Company v Customized Computer Training. Two disputed domain names <stopcompusa.com> and <bancompusa.com> were involved into this case. Although anyone can find without difficulty no any substantial difference between the generic terms “stop” or “ban” in this case and “sucks” in the previous case, the panel provided a distinctive answer by holding that absolutely no any confusion or similarity (much less identity) existed between the two domain names and the trademark of the complainant and no one could confuse “compusa” and <stopcompusa.com> and <bancompusa.com>. This more reasonable decision is apparently inconsistent with the above one, albeit facts and circumstances are similar.
Rights or legitimate interests ---- honest concurrent use in connection with a bona fide offering of goods or services
In real world, the trademark owners can honestly and concurrently use the same mark in different places or in different industries at the same place, with exception that the involved trademark is a famous mark. But on the borderless Internet, the mutual owners of a trademark are generally unwilling or impossible to share a unique domain name corresponding to that trademark. Hence, disputes will arise. A well-known one is the Fuji case. The complainant, Fuji Photo Film, was a famous international company based on Japan, while the respondent, Fuji Publishing Group, is a US company with the domain name <fuji.com> in its hand. Considering facts and evidence, the panel noted that although the complainant had owned trademark rights to “Fuji”, its trademark had not encompassed Internet services provided by the respondent using its corporate name “Fuji” since 1992. This meant that both parties’ concurrent use was honest and accordingly legitimate. On this ground, the respondent was entitled to rights and legitimate interests in its domain name. This case manifests that the prior honest concurrent use (in this case, the respondent’s use of its company name “Fuji” and domain name <fuji.com>) and the rights or legitimate interests arising therefrom are enough to prevent other right-owners from taking away the domain name.
An inharmonious decision has been found in the Madonna case. The complainant in this case was the celebrated American entertainer Madonna who owned common law rights in the mark “Madonna”. However, although the respondent, the holder of the domain name <madonna.com>, had acquired registered trademark rights in “Madonna” in Tunisia in 1998, and although it began to use the domain name for an adult entertainment web site (different use) two years before the notice of the dispute, and although that site contained a notice stating “Madonna.com is not affiliated or endorsed by the Catholic Church, Madonna College, Madonna Hospital or Madonna the singer” , and moreover, although “Madonna” was a common word and had been used by various entities in various ways, the panel firmly asserted that the respondent had lacked rights or legitimate interests in the domain name. Compared with the Fuji case, anyone should envy Madonna for her luck.
Substantially, this panel reached such a controversial conclusion largely by consulting the US standards other than the UDRP itself. Is it wrongful? No, a panel has enough power to do as such because the Rules provide that a panel shall “decide a complaint in accordance with the Policy, these Rules and any rules and principles of law that it deems applicable.” Here, a notable fact is that the panelists of the authorized providers usually come from different countries with different legal systems, rules and principles. For instance, the panelists in the WIPO have come from 42 different countries. In consideration of these, how could one imagine the decisions made by them under various rules and principles but without any inconsistency?!
Rights or legitimate interests ---- legitimate noncommercial or fair use
In the UDRP, no explicit protections exist for the domain name holder’s legitimate use for criticizing the trademark owner, although the holder can generally claim that it is “making a legitimate noncommercial or fair use of the domain name”. Accordingly, the resolution under the UDRP to cases that relate to free speech has been proved inconsistent by the following cases.
A conspicuous case is The New York Times Company v New York Internet Services (NYIS). In this case, the respondent registered the domain name <newyorktimes.com> for the purpose of providing a public forum for expression of opinions and comments regarding complainants publications. To this question, the panel answered that:
“Even if the site at newyorktimes.com were purely a commentary or opinion site, NYIS would not be entitled to use THE NEW YORK TIMES mark in the domain name… even though NYIS may have the right (1) to establish a commentary or opinion site re The New York Times and (2) to mention The New York Times at the site and to reproduce the mark there, NYIS is not necessarily entitled to use The New York Times name or mark as part of NYIS’s address for such a website. A domain name is not only an address, it is also a personal identifier. Many addresses have been available to NYIS which would not in any way impinge on the trademark rights of The New York Times.”
Based on such understanding, the panel concluded that the respondent’s use of the domain name in question had been not legitimate noncommercial or fair use as set out in the UDRP and the name should be transferred.
In contrast, an opposite outcome has been found in the Bridgestone Firestone case. The complainants, three corporations, have registered the trademarks “Bridgestone”, “Firestone” and various permutations thereof; whilst the defendant, a former employee of the complainants, registered the domain name <bridgestone-firestone.net> for criticism and commentary about the complainants. After discussion and findings, the panel concluded that:
“Although the Domain Name registered by Respondent is identical or confusingly similar to the trademarks in which the Complainants have rights, the Respondent has legitimate fair use and free speech rights and interests in respect of the Domain Name, and the Respondent has not registered and used the Domain Name in bad faith. The Panel therefore denies the claim of the Complainants for transfer of the Domain Name.”
Obviously, different panels have different understandings in this area.
As usual, use of a person’s own name as a web site could be a legitimate use, even though there is another person or company that has rights or legitimate interests on it. For example, in the Penguin case, the respondent, Mr. Katz had a well-known nickname “Penguin” for many years and had registered the domain name <penguin.org> as a web site of his family. In this case, the panel unanimously considered it reasonable for a person to register a domain name based on a nickname, for in which that person had legitimate interests. Hence, Mr. Katz was entitled to retain his domain name.
Nevertheless, it doesn’t mean that every panel would like to view such kind of use legitimate. An example is the Oxford case. In this case, the respondent had registered the domain name <oxford-university.com> and asserted that his personal name had just been changed into “Oxford University” too. It was true that the respondent’s new name had not well been supported by evidence. But the panel said that even if the respondent had indeed made a change of name, he still had no rights or legitimate interests in respect of the domain name in question. No any reasons were given by this panel. Such a subjective conclusion reflects the panel’s prejudice against the respondent that has led to the inconsistency of decisions.
Bad faith ---- passive holding
The most popular conduct of bad faith in relation to domain names is cybersquatting. The first decision under the UDRP was just dealing with it. In that case, the panel found that the respondent’s offer to sell the domain name <worldwrestlingfederation.com> to the complainant for valuable consideration in excess of out-of-pocket costs had authenticated his registration and use in bad faith. This case, among others, indicates that the UDRP is sufficient to deal with common forms of cybersquatting. However, there are some particular types of cybersquatting as well, for instance, so-called passive holding or pseudo-cybersquatting.
Pseudo-cybersquatters register domain names that resemble trademarks but never use them, or in other words, they passively hold domain names (i.e. inaction). While the UDRP clearly targets ordinary cybersquatting, its application to passive holding is not so clear. However, paragraphs 4(b)(i), (ii) and (iii) of the UDRP may be deemed possibly relevant to passive holding, on which the panel may rely. For instance, in the Toshiba case, the panel ruled that the respondent had been inactive and in bad faith on several grounds, one of which was the respondent had engaged in a pattern of conduct of registering popular and well-known marks as domain names, which had fallen under paragraph 4(b)(ii) of the UDRP.
Apart from those, are there any other standards or principles which could be used to examine passive holding? A landmark case in this respect is the Telstra case. In this case, the respondent’s domain name <telstra.org> was identical to the complainant's registered trademark, but the respondent had not yet used that domain name for any web sites or any other online presence, and nor offered to sell or otherwise transfer that name. Bearing all of these in mind, the panel concluded that the respondent’s passive holding had amounted to use in bad faith. For this purpose, the panel emphatically added that:
“…what circumstances of inaction (passive holding) other than those identified in paragraphs 4(b)(i), (ii) and (iii) can constitute a domain name being used in bad faith? This question cannot be answered in the abstract ; the question can only be answered in respect of the particular facts of a specific case. That is to say, …, the Administrative Panel must give close attention to all the circumstances of the Respondent’s behaviour. A remedy can be obtained under the Uniform Policy only if those circumstances show that the Respondent’s passive holding amounts to acting in bad faith.”
Hopefully, the “all the circumstances” doctrine originated from this case would act as a helpful supplementary to the UDRP. Unexpectedly, it was followed by two extremes. One extreme is to abuse this doctrine. In the case of Aeroturbine Inc. v Domain Leasing Ltd., the sloppy panel simply asserted that under the circumstances the bad faith registration of the domain name followed by inaction or passive holding had constituted bad faith use of the domain name, in the absence of any proof.
On the contrary, the other extreme is to get around this doctrine and of course paragraphs 4(b)(i), (ii) and (iii) of the UDRP. The Sporoptic Pouilloux S.A. v William H. Wilson case is a typical instance. In this case, the panel emphasized that before a change in the Policy, the case of registration in bad faith by the cybersquatter but “without any form of use in any sense of the word, however deplorable, did not fall under the Policy.” The most funny is that the panel stated that “Complainant did not waste its time by submitting this case… The decision, when served upon the Respondent, may serve as a warning to him not to ‘use’ the domain name in any way… Complainant would be free to submit again… should Respondent start any form of ‘use’ of the domain name in the future.” In fact, this domain name has been sentenced to death by the panel unless the respondent automatically gives it up.
These two extremes not only showed inconsistency of decisions, but further also disclosed the panel’s arbitrariness when applying the UDRP. Ironically, almost no risks will be run by the panel when doing like this.
Bad faith ---- disrupting the business of a “competitor”
To prove the respondent’s bad faith, the complainant may claim that the domain name registration is primarily for the purpose of disrupting the business of a competitor. Since the UDRP has failed to make some detailed explanation on the term “competitor”, one can intuitively understand that a competitor should refer to the complainant who is a commercial rival of the respondent. Simultaneously, one can also think that a competitor needn’t be a commercial opponent of the respondent. For example, in a typo-squatting case, namely Estée Lauder Inc. v estelauder.com, estelauder.net and Jeff Hanna, when dealing with this sticky issue, the panel held that “unlike those of paragraph 4(b)(iv), [paragraph 4(b)(iii)] do not require that a party be motivated by ‘commercial gain’, further supports the fact that the genre of competition referred to need not be commercial. Respondent is, moreover, clearly competing with Complainant for the attention of internet users, which it hopes to divert to its sites.” Bad faith was consequently found on this ground.
But such an inconsistent but plausible interpretation of “competitor” will encounter two problems. First, it has gone so far that any a complainant could be fully eligible for a competitor of the respondent, even without any commercial relationship between them. Therefore, the term “competitor” specifically used in the UDRP seems to be unnecessary. Second, many legitimate users of domain names will be deemed as in bad faith according to such a interpretation. For instance, in the circumstance that a respondent is merely using its domain name to criticize the complainant and of course such criticisms will to some extent disrupt the complainant’s business, if viewing the complainant as the competitor of the respondent, the panel had to logically reach the conclusion that the respondent was using that name in bad faith pursuant to paragraph 4(b)(iii) of the UDRP. This implies that such an interpretation will definitely brand most forms of fair use or free speech as bad faith, which is evidently ridiculous.
Anyway, what in the world is correct or authoritative understanding? The UDRP and its decisions did not provide a clear answer.
Reverse domain name hijacking
As defined in the Rules, reverse domain name hijacking means to use the UDRP in bad faith with intention to deprive a domain name holder of its domain name. The Rules also note that “if after considering the submissions the Panel finds that the complaint was brought in bad faith, for example in an attempt at Reverse Domain Name Hijacking… , the Panel shall declare in its decision that the complaint was brought in bad faith and constitutes an abuse of the administrative proceeding.” Unquestionably, the UDRP has provided unclear guidelines on both whether it is compulsory for a panel to make a finding at the request of the respondent and how to make a finding of reverse domain name hijacking. Thus, the inconsistency among decisions came out, i.e. some panels refused to make a finding and some panels would like to do so. In the Stick case, the panel said that:
“No member of the Panel was familiar with any decisions on this point. On the facts presented, there was a strong feeling that this case may well be an example of reverse domain name hijacking. However, without guidance from ICANN the Panel was reluctant to so rule… the Panel urges ICANN to modify the UDRP to provide the needed guidance, hopefully at some time in the near future.”
In the case of Dog.com, Inc. v Pets.com, Inc., the panel stated more outright that “the Arbitrators decline to make a finding of bad faith or reverse domain name hijacking against the Complainant.”
Some panels tried to make a finding. However, the decisions made by them have led to another inconsistency, i.e. different rules applied.
In the Safari Casino case, after concluding that the respondent’s registration and use of the domain name were not in bad faith, the panel additionally ruled that “as the Respondent alleges, the intent of the Complainant in bringing this action appears to indicate… bad faith intent on the part of Complainant in order to use this procedure to orchestrate a reverse domain name hijacking.” Regrettably, no any reasoning was provided and no any rule employed by this panel in order to draw such a conclusion.
In another case, namely the Breakthrough case, the panel denied the reverse domain name hijacking based on two rationales: First, the complainant owned the trademark rights to its trademark which was identical or confusingly similar to the disputed domain name. Second, the Rules for UDRP had just been adopted a few months before the complaint and therefore “it would be appropriate, in this very particular instance, to give Breakthrough (the Complainant) the benefit of the doubt.” (What a funny reason!)
Fortunately, in the case of Smart Design LLC v Carolyn Hughes, the responsible panel developed a better and practicable rule:
“What is meant by ‘bad faith’ in this context? Clearly, the launching of an unjustifiable Complaint with malice aforethought qualifies, as would the pursuit of a Complaint after the Complainant knew it to be insupportable. This Panel takes the view that ‘bad faith’ in this context extends also to a person who, while maybe not knowing an allegation to be insupportable, makes the allegation reckless as to whether it is supportable or not.”
By virtue of this rule, the panel successfully found that the complainant’s conduct had constituted the reverse domain name hijacking. But unfortunately, this rule has not yet gained broad recognition and compliance. So, inconsistency will go on.
Inconsistency within a single decision rather than between decisions
In quite a lot of decisions under the UDRP, the simplistic decidendi rationes can be easily found. Moreover, the inconsistency within a single decision can occasionally, but not frequently, be found. Such internal inconsistency can be illustrated by a famous case, namely Gordon Sumner, p/k/a Sting v Michael Urvan.
The complainant was a worldwide famous musician, recording and performing artist who had rendered high-quality musical services under his name, trademark and service mark “Sting” since 1978. On the other hand, the respondent, present holder of the domain name <sting.com> had been using the nickname “sting” and “=sting=” publicly on the Internet for more than 8 years. The domain name <sting.com> was registered by respondent in July 1995, approximately 5 years before the dispute was commenced.
In this case, the evidence, provided by the respondent and adopted by the panel, showed that about several years before this dispute the respondent had been using that domain name to point to a web site for email purposes, and had used his nickname “sting” or “=sting=” with global internet gaming services, and had commenced to develop a web site intended to be located at that domain name.
Considering these circumstances, the panel concluded that:
“it is far from inconceivable that there is a plausible legitimate use to which the Respondent could put the domain name. The Respondent has asserted a legitimate use to which he has put, and intends to put, the domain name.”
Based on such an understanding of evidence, the panel ruled that the respondent had not registered and used the domain name in bad faith.
It has been confusing that such an understanding should be reasonably used to sustain the respondent’s rights or legitimate use rather than no bad faith, for it is more relevant to paragraph 4(c) (especially 4(c)(i) and (iii)) than to paragraph 4(b) of the UDRP, albeit two paragraphs have some complicated relations.
However, on the issue of respondent’s rights or legitimate interests in the domain name, the panel ruled that:
“On balance, therefore, this Administrative Panel finds that the Respondent does not have a right to or a legitimate interest in the domain name, in the sense in which that concept is used in paragraph 4(a)(ii) of the Uniform Policy.”
Such a determination is inconsistent with the evidence and the former understanding of the panel, and the whole decision is controversial because of inconsistent decidendi rationes.
Conclusion
From a rough discuss above, some verdicts can be reached as below:
I. There are two possibilities that could result in inconsistency (if any) of decisions under the UDRP: (i) no clear or rigid description in the UDRP for resolving some types of disputes; (ii) the panel’s failing to comply with the UDRP although some requirements therein are clear enough. Unfortunately, both of them have become the reality, which destine the existence of inconsistency of decisions under the UDRP.
II. Although it is statistically consistent in macro-sense, the majority of decisions has clearly expressed the bias toward complainants against domain name holders. In addition to inconsistency in micro-sense, the UDRP is faced with the danger of downfall of authority and acceptability. Anyway, such a worry may be superfluous because the statistics also show that hitherto only 36 decisions under the UDRP have been challenged in courts and it is uncertain whether courts have come to the same conclusions as the UDRP decisions.
III. The UDRP has a formidable dilemma. On the one hand, as an Internet self-regulatory alternative dispute resolution policy other than an international law, for the purpose of worldwide acceptability and related flexibility, it may not formulate detailed provisions and enhanced remedies to challenge national jurisdiction. On the other hand, in order to reconcile decisions, it must provide as clear guidance as possible.
Are there any other ways available to us for harmonizing decisions? Some suggestions, for instance, setting up the appellate procedure, specifying the panel’s liability derived from wrongful arbitration and so on, can serve as considerable solutions for the UDRP to fulfil its potential. However, no any measures have been adopted by ICANN to date.