Distance is not only geographical and temporal but also measured by product or service offerings. The principle underlying temporal distance is that the respondent could not have registered the domain name in bad faith if the trademark had not then come into existence. Subsequent bad faith use (except in the view of panelists who argue for “retrospective” bad faith) is not grounds for forfeiture of the domain name under UDRP. The greater the distance of all three combined the less likely that the complainant can prevail on the bad faith element, which is the ultimate test.

Geographical distance by itself does not favor the respondent’s good faith claim, but temporal distance (acquiring the domain name prior to existence of the trademark) does. The numerous proceedings by western complainants against eastern respondents are more likely to be resolved in complainants’ favor because geographical distance is unpersuasive where the domain model is merely farming the domain name for pay per click revenue. But, add product or service distance and the respondent is a clear winner.

The reverse is that the closer in geographical and product/service distance the more persuasive complainant’s allegations that the respondent lacks rights or legitimate interests and has registered and is using the domain name in bad faith. This is illustrated in Hoboken Publications, LLC v. Hoboken411.com, LLC., FA1111001414530 (Nat. Arb. Forum January 13, 2012). The parties in this proceeding “operat[e] in the same limited Hoboken geographical area and in the same general business as Complainant.” The likelihood is greater that the respondent “must have had actual knowledge of Complainant and its HMAG mark relative to print and online services.” The inference is clear that

In seeking to own the disputed domain name, identical to Complainant’s mark, and attaching a website at the name that generates advertising revenue – the same sort of revenue sought by Complainant – Respondent surely intended to divert such revenue from Complainant (i.e., disrupt Complainant’s business) and gain commercially from the likelihood of Internet user confusion with respect to the source, sponsorship, affiliation or endorsement of Respondent’s website.

In many instances close geographical distance make denial of knowledge implausible – Layby Services Australia Pty Ltd. v. Chrisco Hampers Australia Ltd., D2009-1066 (WIPO November 3, 2009) () (Direct competitor who registered the disputed domain name one year after Complainant established its business) – while far distance supports the opposite inference, namely that more likely than not there is neither knowledge nor targeting. EURO DATA GmbH & Co. KG v. Excel Signs, D2009-0465 (WIPO May 5, 2009) (Germany and U.S.A) contesting a nondescript trademark EURO DATA (). As the distance increases actual knowledge is more likely to be found only where the trademark in issue is particularly strong and heavily advertised or promoted.

An interesting footnote on the heavily advertised or promoted is international companies with strong presences in respondents’ markets. Although there is great geographical distance in one respect (corporate headquarters), it is local in another (operating a business). La Quinta Worldwide, L.L.C v. Rudi Seiberlich, D2011-0385 (WIPO April 11, 2011) (Complainant “is the registered proprietor of many trademarks for LA QUINTA and variants of that trademark both within the United States and internationally, including in Italy where the Respondent is domiciled).

The burden of proving a violation of the Policy is lighter for suggestive and arbitrary and heavier for generic and descriptive terms current in commercial activity – HSM Argentina S.A. v. Vertical Axis, Inc, D2007-0017 (WIPO May 1, 2007) in which the Panel pointed out that eEvidence of third party use of terms identical to the trademark “mitigates against a finding that the Respondent knew specifically of the Complainant or its mark when the disputed domain name was registered.” Denying knowledge is unpersuasive where the respondent uses the domain name to advertise the same goods or services as the complainant. A case in point is ChemRite CoPac, Inc. v. Isaac Goldstein, D2010-0279 (WIPO May 7, 2010) (Parties in the United States and Hong Kong). The Respondent denied knowledge, but populated its website with competitive links.

Use speaks for itself. So, the Panel in ChemRite transferred the domain name to Complainant. If no other inference can be drawn except that the respondent is taking advantage of the complainant’s trademark, respondent loses. Hoboken Publications wins for the same reasons.

Gmlevine

The test for establishing confusing similarity is relatively modest. Low though the bar is set, however, it is not satisfied by simply showing that the trademark and the domain name bear a similarity of parts. “The concept of ‘confusingly similar’ lies at the very heart of all trademark conflicts, where the test is whether the names are phonetically, visually or conceptually similar.” Advance Magazine Publishers v. MSA, Inc. & Moniker Privacy Servs., D2007-1743 (WIPO Feb. 28, 2008). In assessing domain names under the first element of the Policy panelists apply rules familiar to trademark infringement. Put another way, Todito.com, S.A. de C.V. v. Affordable Webhosting, Inc., D2004-0688 (Nat. Arb. Forum Oct. 11, 2004):

when determining whether . . . similarity may give rise to situations of confusion or false association, it is necessary to consider not only the degree of similarity of the terms in the strict sense but also the degree of distinctiveness of the mark given that, obviously, the greater the distinctiveness or originality of the registered mark, the greater the risk of confusion.

A good summary of views is set out in salesforce.com, inc. v. Domain Admin / Internet Venture Holdings, Inc. (ivh.com), FA111100 1416951 (Nat. Arb. Forum January 5, 2012). Letters and words added to or subtracted from the trademark are evidence of confusing similarity if the similarity of a part suggests the whole, but not confusingly similar if the whole contains its own meaning and the domain name is used consistently with it. Letters and words added to or subtracted from the trademark are evidence of confusing similarity if the similarity of a part suggests the whole.

Thus, the mere addition of a letter, or the creation of similar word, can create a domain name that is confusingly similar to a mark. Either the variation is phonetically identical to the mark or it renders the domain name indistinguishable from it. To illustrate the first example, substituting a “k” for the letter “c” does not change the pronunciation of the word, as in ECOLAB and . Ecolab USA Inc. v. Tomasz Kluz / Ekolab s.c. Tomasz I Aleksandra Kluz, FA 1386906 (June 3, 2011). The second example is WIRED and <wwwwired.com> where “wired” is the dominant element of the mark. Advance. In Ecrush, Inc. v. Cox, Davis & Simpson, LLC, D2004-0552 (WIPO Sept. 11, 2004) (ESPIN THE BOTTLE and) the Respondent changeda single letter, “i” for “e.” The Panel found that the “one-letter change is not a misspelling of Complainant’s mark that results in an otherwise nonsensical word or phrase; on the contrary, the Disputed Domain Name and Respondent’s other domain names are everyday phrases incorporating the name of the parlor game.”

“These cases” (the Panel in salesforce.com explains)

are in contrast to the present case. First, the word “forces” has an apparent relationship to the military, thus having no evident connection to any business activity of the Complainant. Second, as to SFORCE and FORCE.COM there is no phonetic or visual similarity between the domain name and Complainant’s marks. Although the word “forces” is arguably visually similar to Complainant’s FORCE mark, the lack of distinctiveness to the mark, and the demonstrably different meaning of the words, results in a lack of any confusing similarity between the two.

The dissenter in Open Society Institute v. Gil Citro, FA1007001333304 (Nat. Arb. Forum August 24, 2010) – who is the sole panelist in salesforce.com – proposed that the similarity must be confusing to an “objective bystander.” Applying that standard he rejected <opensociety.org> as being confusingly similar to OPEN SOCIETY INSTITUTE). However, the “objective bystander” standard carries a burden of subjectivity. For this reason, Panels generally tend to give complainant the benefit of the doubt on this element. In salesforce.com the objective bystander (as fictively breathed into life by the Panel) is called upon to make an assessment whether is confusingly similar to FORCE. Framed by the Panel in Advance, the question is whether the “alphanumeric string comprising the challenged domain name is identical to Complainant’s mark or sufficiently approximates it, visually or phonetically, so that the domain name is ‘confusingly similar’ to the mark.” The domain name trademark FORCE

Whatever similarity in part between trademark and domain name does not prove confusing similarity to the whole.

Gmlevine

Unless personal names have achieved trademark status they are not protected under the UDRP. Report of the Second WIPO Internet Domain Name Process, The Recognition of Rights and the Use of Names In the Internet Domain Name System, dated September 3, 2001 which states: “the application of the UDRP to the protection of personal names [is authorized only] when they constitute trademarks,” Paragraph 179. Excluded are living persons whose reputations are earned out of the spotlight of commerce even though active in the figurative scrum of their businesses or professions. This rule applies however extraordinary the individual’s contribution is to business, science, politics, the professions and the academy. The WIPO Report states that it may result in an “injustice” and is undoubtedly an unhappy limitation: “many sensitivities [will be] offended by the unauthorized registration of personal names as domain names” and the “result is that there are some perceived injustices.” Report paragraph 199.

In contrast, the Lanham Act provides a remedy for cyberpiracy of personal names that do not qualify for trademark protection, formerly codified at 15 U.S.C. 1129, now 15 U.S.C. 8131 (Cyberpiracy protections for individuals). Explaining the necessity for the provision, Senator Hatch, noted

As with trademark cybersquatting, cybersquatting of personal names poses similar threats to consumers and e-commerce in that it causes confusion as to the source or sponsorship of goods or services, including confusion as to the sponsorship or affiliation of websites bearing individuals’ names. In addition, more and more people are being harmed by people who register other people[’]s names and hold them out for sale for huge sums or money or use them for various nefarious purposes.

There has been a couple of reported cases and several unreported. In one of the reported case the court granted an injunction, Schmidheiny v. Weber, 319 F.3d 581, 66 U.S.P.Q.2d 1062 (3d Cir. 2003), and in the other it denied it, Carl v. BernardJCarl.com, 662 F. Supp. 2d 487 (E.D. Va. 2009), aff’d 409 F. App’x 628, 630 (4th Cir. 2010) (per curium) (unpublished). In Schmidheiny, the court based its conclusion on a consideration of factors including the defendants’ past similar behavior, the short time-lapse between registration and offer for sale, and the placement of the domain names “for auction at a price that had no reasonable connection to its value to anyone other than the plaintiff.” 319 F.3d at 628.

In one of the unreported case from the Middle District of Florida, Salle v. Meadows, 6:07-cv-1089-Orl-31 (August 6, 2007) the court granted a preliminary injunction against defendant who admitted purchasing the domain name and attempting to sell it to the plaintiff for $9500. He claimed he was merely attempting to recover money plaintiff owed him. The court rejected this “defense”: “cyber-extortion is not a permissible way of recovering a debt” and ordered the domain name transferred.

The latest of these cases, unreported from New York has just recently been filed, Paul Bogoni v. Vicdania Gomez, 11 Civ. 08093 (S.D.N.Y. December 28, 2011) the court has also granted an injunction. The court found that “defendant purchased [two domain names] – multiple versions of the plaintiff’s real name – for less than twenty dollars in total, and, within several days, posted an offer to sell the Domain Names for $1,000,000 each.” The court concluded that plaintiff stated a claim under § 8131(A) and proceeded to examine whether defendant’s defense under § 8131(B) “remove[d] her from the reach of the ACPA.”

Section 8131(B) requires defendant to prove absence of bad faith. This she was unable to do. The Section reads:

A person who in good faith registers a domain name consisting of the name of another living person, or a name substantially or confusingly similar thereto, shall not be liable under the paragraph if such name is used in, affiliated with, or related to a work of authorship protected under Title 17, including a work made for hire, as defined in section 101 of Title 17, and if the person registering the domain name is the copyright owner or licensee of the work, the person intends to sell the domain name in conjunction with the lawful exploitation of the work, and such registration is not prohibited by contract between the registrant and the named person.

The defendant concocted a set of facts that made it appear as though there was a copyright issue. The court concluded is was a sham, that the registration violated the statute and granted a preliminary injunction enjoining defendant from continuous use of the domain names.

Gmlevine

Complainants of trademarks on the lower end of the classification scale have a greater burden to prove bad faith registration even where respondent is in the domain name business and there is proof of temporary infringing use of the domain name. “The number of other persons or entities holding identical if non-competing marks and the number of other users with rights in [a] name are clear evidence of the limited ownership claims of the Complainant.” Electronic Arts Inc. v. Abstract Holdings International LTD / Sherene Blackett, FA1111001415905 (Nat. Arb. Forum January 4, 2012) disputing over <ssx.com>. The Respondent in this case is a high volume registrant. It acquired the disputed domain name as part of a portfolio of three and four letter generic names from a prior holder.

High volume registrants have an enhanced duty of care, but it does not impose an impossible burden of investigating every domain name for possible infringement of a third party’s trademark rights. Where Panels have rejected respondent’s good faith allegations it generally involves trademarks well-known in the market place. For registrants who regularly engage in the business of registering and reselling domain names, and/or using them to display advertising links they must be able to show that

1. It had made good faith efforts to avoid registering and using domain names that are identical or confusingly similar to marks held by others;
2. The domain name in question is a dictionary word or a generic or descriptive phrase;
3. The domain name is not identical or confusingly similar to a famous or distinctive trademark; and
4. There is no evidence that the Respondent had actual knowledge of the Complainant’s mark.

Where there is no evidence of bad faith registration the Policy does not penalize a respondent for changing use of the domain name after notice that it contains infringing content. Inadvertent transgressors typically involving domain names composed of generic or descriptive words and combinations used in their commonplace sense are treated more kindly than respondents advertently taking advantage of the complainant or its trademark. Enhanced investigation of domain names identical or confusingly similar to trademarks can be traced to Mobile Communication Service Inc. v. WebReg, RN., D2005-1304 (WIPO February 24, 2006) contending for. The Panel in that case held that MOBILCOM (an unregistered trademark) was not a dictionary word or a generic or descriptive phrase. He concluded that where

a respondent registers large swaths of domain names for resale, often through automated programs that snap up domain names as they become available, with no attention whatsoever to whether they may be identical to trademarks, such practices may well support a finding that respondent is engaged in a pattern of conduct that deprives trademark owners of the ability to register domain names reflecting their marks.

This is in contrast to high volume registrants such as the respondent in Electronic Arts who is “in the business of buying and selling generic domain names.”SSX is composed of generic letters. “Complainant clearly does not have an exclusive monopoly” on letters that can equally and legitimately be employed by other parties in unrelated fields. Although it is not a defense that respondent’s hosting company was responsible for placing content of a website, nevertheless where hyperlinks are quickly removed it is some evidence of its intentions. “Respondent alleges that the resolving website was only supposed to contain generic hyperlinks and not competing hyperlinks.” The Panel accepted this explanation. “Generic hyperlinks” is the broadest generic meaning that an ordinary dictionary word is capable of carrying. Tomsten Inc. v. Registrant [7281], FA 925448 (Nat. Arb. Forum April 20, 2007). For made up words like “mobilcom” there can be no legitimate use by another party in an unrelated field. For “ssx” the opposite is true.

Dissents in a number of cases have argued for greater diligence. The dissent in BAWAG P.S.K. Bank für Arbeit und Wirtschaft und Österreichische Postsparkasse Aktiengesellschaft v. Future Media Architects, Inc., D2006-0534 (WIPO July 28, 2006) for example suggested that registrants who use their vast accumulations to exploit a business opportunity have “a greater duty … to verify that the use of a domain name … is legitimate.” In that case the parties were contending for. But, the consensus view is that bulk registrants may legitimately hold generic domain names absent clear and convincing evidence that their use demonstrates knowledge and targeting of the trademark.

Unintentional hyperlinking quickly removed upon receipt of notice reinforces a respondent’s assertion of good faith registration. By “[m]odifying such a website also may attenuate the harm to a complainant whose trademark or commercial interests are allegedly compromised by the website prior to remediation.” American Airlines v. James Manley d/b/a Webtoast Internet Services, Inc., FA1006001330044 (Nat. Arb. Forum July 27, 2010)(). The Panel’s point is that respondents (in the inadvertent class) should not be discouraged from “this positive behavior…. [T]he fact of a website’s remediation should not create an evidentiary inference adverse to the respondent.”

Gmlevine

Lego is one of those trademarks instantly recognizable in the marketplace. The number of UDRP proceedings against Lego infringers runs into the dozens (I’m probably underestimating). Here is a baker’s dozen decided in December 2011: <legodiscounter.com>, <legoninjagofiretemple.com>, <legoverhuur.com>, <legoland florida.com>, <legolandorlando.com>, <legosoft.com>, <legowatches.org>, <lego-fan.net>, <legomen.com>, <legostarwarsgames.org>, <legoharrypotterquidditch match.net>, <legoarmory.com>, <legocitycorner.com>. How can a respondent escape losing a “lego” domain name? The issue in the first instance is how the domain is being used and the purpose for which it was registered. There has to be a match. Even with famous trademarks, complainant can be defeated if the purpose for the domain name is “legitimate noncommmercial” and its use fair under paragraph 4(c)(iii) of the Policy. Complainant was unsuccessful for <legoworkshop.com> in LEGO Juris A/S v. Domain Administrator, Matthew Griffith, D2011-1263 (WIPO October 1, 2011). Why?

The Panel in Velcro Industries B. V. and Velcro USA Inc. v. allinhosting.com/Andres Chavez, D2008-0864 (WIPO July 28, 2008) explained that “it would take an exceptional case to succeed where there was no malicious or exploitative intent directed at the complainant [or its trademark] at time of registration of the domain name.” The Respondent in Velcro was a creative “velcro” artist. In Lego v. Griffith, the Respondent stated that he

registered [the domain name] on behalf of his 15 year old 10th grade son who has been a LEGO enthusiast since he was 3 years old. His intent was to create a website that showed off the creations that he (the son) has built over the years in “both photographs and occasional video while also showing steps to build his creations”.

It was “part of a high school project” and that his son has

chosen this to be his project as it allows his passion for the product, the creativity of building using the product and the programming technology required to create and maintain the site to flourish while helping other kids to build creations that they might not be able to dream up on their own and to share building ideas”.

A 4(c)(iii) defense is credible if the use matches the assertion of noncommercial or fair. In a 2008 case Respondent argued that his registration of <narnia.mobi> was a birthday present for his son who was a fan of the Narnia books, C.S. Lewis (PTE.) Ltd. v. Richard Saville-Smith, D2008-0821 (WIPO July 21, 2008). His sole intent of using the domain name for a child’s e-mail address and with no commercial motivation establishes rights or legitimate interests under the Policy. But, his use (or, rather nonuse) of the domain name gave him away:

the Panel cannot envision any plausible, good faith basis upon which the Respondent could have concluded that he was free to appropriate the Complainant’s distinctive and widely known NARNIA mark for use as a personal email address.

Contrastingly, in the Lego case there was a matching of allegations with use of the website. The Panel “is confirmed in its view that the Respondent has shown that the circumstances specified in paragraph 4(a)(ii) exist. He continues,

In many circumstances, bad faith registration in the face of such knowledge can be readily inferred. However, the record here would suggest that, notwithstanding such knowledge the Respondent and his son did not intend to cause any confusion or diversion of custom and did not, indeed, believe this to be likely, given the nature of their intended website. While the facts of each case will differ, in the present case the Panel accepts that this is sufficient to dispel the inference of bad faith registration that might otherwise arise.

The language of 4(c)(iii) “makes it clear that the operator of a noncommercial fan site can take advantage of this provision unless the operator, with an intent for commercial gain, is misleadingly diverting consumers or tarnishing the mark.” Estate of Francis Newton Souza v. ZWYX.org Ltd, D2007-0221 (WIPO June 18, 2007).

Gmlevine

There have been a number of cases lately in which complainant has either failed to prove a prima facie case or been unable to overcome respondent’s evidence of its right or legitimate interest in the disputed domain name. Respondent’s use of the domain name is a critical factor, but so too is the trademark. Longevity and recognition in the marketplace cannot trump a respondent’s right or legitimate interest unless it fails to offer affirmative evidence of its choice.

A respondent’s loss of its trademark for failure to file a declaration of use between the 5th and 6th years of registration is not fatal to either its right or legitimate interest in a domain name. Stonz Wear, Inc. v. Framez l-Wear, D2011-1764 (WIPO December 8, 2011). There is no rule that a respondent must own a trademark; indeed, it is more generally true that a respondent is not nor ever was a trademark holder, but that its right or legitimately interest arises either by its use of the domain name [paragraphs 4(c)(i) and (iii)] or through the coincidence of its business or personal name with the domain name [paragraph 4(c)(ii)]. Complainant’s prima facie burden on the second element is only satisfied by alleging that it has not given permission for respondent to register the domain name and that the respondent is not “commonly known” by the domain name. To allege one without the other does not satisfy the burden. In Stonz, the Respondent operated a business years prior to the Complainant acquiring its trademark.

In The California Milk Processor Board v. Center Ring Productions, LLC., D2011-1689 (WIPO December 1, 2011) (see last week’s post for another California Milk dismissal) the Respondent is a corporation owned by Mr. John Milke (pronounced with two syllables). Complainant challenged the registration of. GOT MILK? is “extremely well known in the United States.” Nevertheless, the Panel found it “difficult to conceive of a better means of establishing that one is ‘commonly known’ by a word than by showing that the word is his family name. Mr. Milke has been commonly known by that word all his life.” It does not explain the choice of prefix, but here the Panel disregarded “got” as a ” mere” prefix, implying that it undercut Respondent’s legitimate interest. In the Panel’s “assessment [it did not] take the Respondent outside the safe harbor of paragraph 4(c)(ii).” The missing ingredient for the Complainant is evidence of commercial use such that it could be argued that the prefix was not just “mere” but a subterfuge for taking advantage of the trademark. Had it been otherwise “there might be an argument that there needed to be proof that Mr. Milke had been commonly known by the entire phrase” but “Respondent’s use of his surname as the dominant feature of a non-commercial website keeps him within paragraph 4(a)(ii)’s safe harbor.”

A coda in California Milk on reverse domain name hijacking is interesting in pouncing on the Complainant for not recognizing that Respondent’s surname undercut the assertion that he lacked rights or legitimate interests in the domain name. The coda in full reads:

After more than a decade of precedent in UDRP proceedings, this Panel believes that “any complainant, and even more so any professional representative of a complainant, should be at least minimally versed in the Policy, the Rules, their scope, and their limits. It is no excuse that a party or its representative is unfamiliar with clear Policy precedent, much less the clear language of the Policy and the Rules themselves”…. So he finds it troubling that the Complainant, represented by counsel, so casually elides an express Policy provision, paragraph 4(c)(ii). In the Panel’s opinion this Complaint would not pass muster under Rule 11 of the United States Federal Rules of Civil Procedure, which treats counsel’s signature on a pleading as his certification, “formed after an inquiry reasonable in the circumstances “that the claims, defenses, and other legal contentions therein are warranted by existing law or by a nonfrivolous argument for the extension, modification, or reversal of existing law . . .” Similarly it is inconsistent with the Complainant’s (or its representative’s) certification, required by paragraph 3(b)(xiv) of the Rules, that “that the information contained in this Complaint is to the best of Complainant’s knowledge complete and accurate, . . .”

This having been said, and although the Panel found that “the Complaint ‘was brought primarily to harass the domain-name holder,’ Rules, paragraph 15(e),” he elected not to enter an order for reverse domain name hijacking. Instead, he chose to “admonish[] the Complainant’s counsel that in future cases he should do his homework before filing.”

Gmlevine

Giving a party the benefit of the doubt is really shorthand for insufficiency of the other’s evidence. In the first instance, complainant has the burden for establishing its case for all three elements, but as it proceeds from element to element its burden grows heavier. Proof that the domain name is identical or confusingly similar to a trademark in which it has a right is a low bar. Uncertainty whether a domain name is confusingly similar or similar but not confusing generally favors the complainant, particularly if the record favors the respondent. Element two, that respondent lacks rights or legitimate interests is heavier but complainant only has to establish a prima facie case, which is essentially a showing that there is no evidence to the contrary. The burden increases if respondent appears and offers evidence that complainant must then refute. Supplementing the record is permitted conditionally. The condition being that it is not cumulative. The heaviest burden is imposed on the third element, that the domain name was registered and is being used in bad faith.

When Panels struggle with this question it is more likely than not that doubt will be resolved in respondent’s favor. In The California Milk Processor Board of San Clemente v. Ryan Leonard, D2011-1665 (WIPO December 5, 2011) contesting the Panel’s final comment is telling:

the Panel wishes to note that while it is to some extent, giving the Respondent the benefit of the doubt with respect to its claimed intended use of the disputed domain name, if in the future such use would not materialize, or the use would change in such a way to suggest a clear abusive intent, this may well constitute a relevant consideration in any future dispute brought over the disputed domain name.

This coda is a warning to the Respondent. If the disputed domain name is used as the Respondent alleges it succeeds because it confirms good faith registration and no bad faith use. But, the domain name would have no value for any use other than the business it describes, namely the selling of “Get Milk Ads.” In this particular case, the Complainant itself identifies such selling as a legitimate business model. “The Complainant further contends that the fame of the GOT MILK? trademark is such that the Complainant’s advertisements are sold as memorabilia on eBay.”

This is exactly what the Respondent has in mind. He (in fact a “he” and a lawyer) “avers that he registered the disputed domain name with the intent to develop a website for the purpose of selling GOT MILK? advertisements, but has yet to make such use of the disputed domain name because of his busy law practice.” He is not sufficiently persuasive in affirmatively demonstrating a right or legitimate interest and hence the Panel’s struggle. “While the Panel does not consider the examples provided in paragraph 4(c) of the Policy necessarily to be exhaustive of the ways in which a respondent may demonstrate rights or legitimate interests, reliance solely on intent to use the disputed domain name in connection with a bona fide offering of goods and services is not sufficient to confer rights or legitimate interests under the Policy.” The reason for this is that to pass muster for a paragraph 4(c)(i) defense, respondent must “demonstrable preparations.” Complainant therefore succeeds on the second element because the Respondent fails to rebut the prima facie case against it.

The existence of a legitimate business in Get Milk ads makes Complainant’s burden heavier to prove registration in bad faith. This is where the nominative fair use is applied:

the issue in this administrative proceeding is not whether the Respondent (or others) has a right per se to sell (or resell) the Complainant’s GOT MILK advertisements. Rather, the issue is whether the Respondent consistent with the Policy may register and use a domain name incorporating the Complainant’s trademark for that purpose. In that regard, the consensus view of UDRP panelists is that a reseller or distributor making a bona fide offering of trademarked goods or services using a domain name which contains the trademark has a legitimate interest in the disputed domain name.

Absent proof that Respondent intended to take advantage of the Complainant’s trademark, bad faith is a maybe and maybe does not suffice to establish bad faith. With a bow to Telstra Corporation Limited v. Nuclear Marshmallows, D2000-0003 (WIPO February 18, 2000), in this particular case “it is … possible to conceive of [a] plausible actual or contemplated active use of the domain name by the Respondent that would … be illegitimate.” But woe betide if Respondent does not follow through with its plan and uses the domain name in bad faith. It would be a “relevant consideration in any future dispute”, meaning that it would support a return trip to a later UDRP proceeding.

Gmlevine

A UDRP record is similar in some respects to a motion for summary judgment in a civil action in that each party puts forward its best case in support and opposition to the complaint. The difference is that in a civil court action a case can proceed to discovery and trial but in a UDRP proceeding the record is all there is. Credibility plays a role just as silences do. Persuasion is partly in the manner in which evidence is presented, but more importantly it is the submission or withholding of evidence that supports a party’s argument. So, offering and withholding evidence cannot fail to impress the Panel. In Fiji Rugby Union v. Webmasters Limited, D2003-0643 (WIPO December 24, 2003), for example, the majority stated that although the case was “finely balanced” it was able to “draw inferences from the manner in which the Respondent has argued its case, retreated from formal statements of fact and generally been less than forthcoming.”

These evidentiary issues bubble up in Case Management Professional Staffing Solutions, Inc. v. Genius Genius of NY Inc., FA1111001414200 (Nat. Arb. Forum December 12, 2011). The Panel found that Complainant had established a prima facie case, thus shifting the burden to Respondent to rebut evidence that it lacked rights or legitimate interests in the disputed domain names which consist of a series of letters, and. The “ceu” component is “continuing education units.” Complainant is a certified CEU provider. “Complainant [alleged that it] has rights in the mark AAACEUS.COM through a trademark registration of that mark with the United States Patent and Trademark Office (“USPTO”) (Reg. No. 3,582,290 filed November 29, 2007; registered March 3, 2009). Complainant also submits that it has common law rights in the mark AAACEUS.COM and that they date back to October 24, 2005.”

The disputed domain names in Case Management differ from the trademark by omitting the final “s” of the second level domain. Respondent countered that the “aaa” series stands for “Above All Academics Inc.” He did not register the domain names in that corporate name because “Complainant delayed Respondent from launching its business” thereby appearing to raise a business dispute.

The Panel observes that Respondent “seeks to [rebut Complainant’s prima facie proof] by advancing an argument, to which the Panel has given careful attention , drawing on the relationship between the parties and Respondent’s version of the history of the registration of the domain names, their use since then and other aspects of Respondent’s business.” In this respect,

Respondent’s case raises the difficult question of whether and how a UDRP panel should approach such disputes where there is a conflict of testimony.

“Undoubtedly,” (the Panel continues)

there are cases where the dispute is complex and probably not appropriate to be resolved by the limited processes available in UDRP proceedings.

This would be the summary judgment result where there is a genuine issue of material, therefore triable fact. However, the Panel continues

On the other hand, a panel has a clear responsibility to make a decision where there is adequate material before it and where it is possible to decide if one party or the other has the better case. In the present case, the Panel has decided that although the parties are in dispute, the dispute is essentially about the domain names and there is sufficient evidence for the Panel to reach a conclusion.

As part of its case, “Respondent denies that it copied Complainant’s trademark and then tried to cover its tracks by making a slight spelling alteration; in other words, Respondent denies that this is a case of typosquatting.” But, “was it typosquatting?” While these issues are clearly “difficult to decide” the Panel concluded that “on the balance of probabilities” is was an example of typosquatting. The reasoning (which is the part of the decision I want particularly to stress) is

Whether Complainant was pre-eminent in its field or not, it is clear that it was in business under its trademark, domain name and website and that it had maintained that position for a period of at least some 18 months, from October 24, 2005 until April 4, 2007, when Respondent registered the first of the two domain names…. It must also be remembered that Complainant had the domain name <aaaceus.com> from November 29, 2004 and that it was registered in the name of AAACEUS.COM. Thus, although it is true that Complainant subsequently applied for a trademark in the same name and claimed that its first use in commerce was on October 24, 2005, its domain name was registered almost a year earlier.

Moreover,

a search using the Wayback Machine atshows that the domain name was actively used and that the first website under that name was on the internet by January 7, 2005, showing a website substantially the same as it remained and as it is today. Not only does this show that Complainant was actively in business during the whole of the time from January 7, 2005 and on the public record as doing so, but that this was the case immediately prior to the two dates when Respondent registered the two disputed domain names.

The rebuttal evidence was not persuasive. The fact that the domain names were actually registered in the name of the present Respondent and not the corporation “make it more likely than not that in registering the two disputed domain names, Respondent was influenced by Complainant’ trademark, that it was in effect copying Complainant’s name and that there was no other legitimate reason for registering the two disputed domain names.”

Gmlevine

Panels are typically faced with a cascade of options in resolving UDRP disputes. Complainants do not necessarily prevail for proving trademark rights and respondents do not lose for failing to appear and defend themselves. First for some perspective. WIPO prepared an informative overview of the UDRP and its role in a paper dated August 2011 “for the purpose of informing discussion at the September 2011 INTA Conference on Trademarks and the Internet on the issue of whether it would be wise to review and make changes to the UDRP, at least at this time.” We learn that over 35,000 cases have been processed by ICANN Providers since 1999 of which the two most productive are WIPO and National Arbitration Forum. The “transfer rate” is slightly lower for WIPO (86.1%) and higher for Nat. Arb. Forum (87.8%).

Respondents either appear and state their cases or they do not. Reflecting only WIPO’s experience in resolving disputes (it decides approximately 52% of cases and Nat. Arb. Forum approximately 45%), appearing respondents prevail in about 30% of the cases with single member panels and 56.7% with 3 member panels. Defaulting respondents (approximately 80% in 2011) lose 93% of the time decided by single member panels and 83.8% decided by 3 member panels. That 16% of defaulting respondents before 3 member Panels retain their disputed domain names is a tribute to panelists’ careful assessment of the record, but is also the result of the conjunctive model of the UDRP.

Not incorporated into the statistics are situations describing circumstances under which complainants lose. Unscientifically, a small percentage of complainants fail in proving trademark rights (the first element of UDRP). Larger but not significant percentages fail on the prima facie case (second element) and a larger and significant percentage fail in proving conjunctive bad faith (third element).

Proving conjunctive bad faith is increasingly difficult the longer the temporal distance between domain name registration and trademark acquisition and less so as the distance narrows. The conjunctive bad faith model is peculiar to UDRP. a number of country code policies and the Anticybersquatting Consumer Protection Act (ACPA) opted for a disjunctive model. An interesting decision from India has just been reported that draws attention to distinctions between UDRP and INDRP in which the respondent (petitioner in the civil proceeding) was a bulk registrant, Koenig v. Arbitrator, O.M.P 132 of 2007 (Delhi High Court, December 14, 2011), but then bulk registrants are treated differently than single registrants. The INDRP follows the disjunctive model.

On the debate over conjunctive vs. disjunctive bad faith WIPO indicates in the Paper that it is not opposed to a change from “and” to “or.” The Paper states that

it is a proposal likely to be attractive to brand owners. There are clear and workable precedents in the ccTLD space (including the Nominet DRP and the .auDRP models). Changing the “and” to an “or” in the UDRP would require a Policy and Rules change (with the above-mentioned risks if pursued through a minority IP ICANN process at this time). WIPO believes this could be a good change for the UDRP in principle, but the timing and process would have to be right.

Interestingly, WIPO introduces this conclusion with the statement that it “bears reflecting that this issue in practice in fact only concerns a very small number of decided UDRP cases overall (findings of clear bad use absent findings of actual or inferred bad faith registration are rare). Whether this assessment is correct, which I think not, the reason why the proposal would “likely … be attractive to brand owners” is that it removes an impediment in the proceedings, namely situations in which domain names registered prior to trademark acquisitions are subsequently used to take advantage of the emerging reputations of trademark owners. One result would be clear by a change in policy and that is that the transfer rate will surely increase.

Gmlevine

The Policy’s scope is limited to “‘abusive registrations’ made with bad faith intent to profit commercially from others’ trademarks.” Second Staff Report on Implementation Documents for the Uniform Dispute Resolution Policy, 4.1(c). The Report continues that “the adopted policy leaves the resolution of disputes to the courts (or arbitrators where agreed by the parties) and calls for registrars not to disturb a registration until those courts decide.” Where there is no intent “to profit commercially from others’ trademarks” there can be no bad faith. Claims for breaches of contract and fiduciary duty, for example, that require the Panel to delve into the parties’ legal relations are generally outside the scope of the Policy.

In Vans, Inc. v. Military and Rescue Supply, FA1109001407743 (Nat. Arb. Forum December 6, 2011), the Respondent is one of a number of affiliated companies one of which either has (or had) a distribution agreement with the Complainant. “Respondent, through its affiliated entities had a long-standing business relationship with the Complainant. Pursuant to that relationship, these affiliated companies sold Complainant’s goods. Although Respondent does not have a distributorship agreement with Complainant, it has continued to sell Complainant’s goods, and other merchandise that does not compete with Complainant’s business.” On these facts, the Panel determined that the disputed domain name was registered in good faith, and since the merchandise on the website to which the domain name resolves is not competitive with Complainant’s business there is no evidence of bad faith use.

As in Vans, the parties in Prevost Car Inc. and Volvo Group Canada Inc. v. Lori Cooper, D2011-1427 (WIPO November 22, 2011) were subject to U.S. jurisdiction and both Respondents were in the business of selling Complainant’s goods. The Panel noted in Prevost that the “claim that the mere use of the domain name itself creates ‘initial interest confusion’ is no longer good law throughout the U.S. under Toyota Motor Sales, U.S.A., Inc. v. Tabari, 610 F.3d 1171 (9th Cir. 2010). “Complainants acknowledge that, at least as of 2004, Prevost Car Inc. expressly granted to Phil Cooper Motor Homes the right to ‘continue its use’ of the <prevost.com> domain name,” but argue that “that consent agreement was ‘revocable’ and that it has been effectively revoked.”

While express revocation would support complainant’s contention, equivocal revocation does not. Thus, in Prevost, “while the unsigned document provided by Complainants evidences its consent in 2004 to the registration and use of the domain name, it does not establish that the consent was revocable, much less unilaterally revocable.” The consensus view is that “there are circumstances where a reseller or distributor may make a bona fide offering of goods and services and thus acquire a legitimate interest in the domain name,” citing WIPO Overview, Question 2.3. “Those circumstances generally include: (1) actually offering the goods and services at issue, (2) using the site to sell only the trademarked goods, and (3) the site’s accurately disclosing or disclaiming the registrant’s relationship with the trademark owner.”

Even though in Prevost, Respondent does not use the site to sell PREVOST vehicles exclusively, the evidence suggests that the site is used primarily to sell PREVOST vehicles and that “that use has for many years been with Complainants’ express consent.” Thus, “[u]nder the particular circumstances of this case, the Panel would find that even if Complainants’ had made a prima facie showing as to paragraph 4(c)(i) of the Policy, that that showing had been rebutted.” Long time use with the complainant’s blessing creates a factual issue beyond the scope of the Policy. “[T]he Panel views the question of the parties’ contractual rights and obligations, if any, relating to the consent ‘agreement’, to be well beyond the scope of these proceedings.”

Gmlevine