This last article on the four new gTLD objections will look at the Legal Rights Objection ("LRO"). While other articles in this series have touched on trademark concepts at certain points (see part one, two and three), issues from that area of the law predominate in LRO. Here we review the pertinent LRO-related trademark concepts, with which many readers likely will have some familiarity from working with domains and the UDRP. Still, the theme of the first three articles applies here: Potential objections are more involved and complicated than they may seem, and require careful thought if they are to be made.
Standing:
Analyzing any new gTLD objection always begins with standing — namely, whether the objector has the right to raise a claim in the first place. AGB 3.2.2. The LRO does not have standing requirements nearly as strict as those for community or as wide-open as those for limited public interest objections, but does require an objector to establish its status as a "rightsholder." While there is not explicit designation of the exact "rights" needed for standing, the Guidebook's later section on the objection standards clearly refers to trademark — i.e., a name, brand, term or characteristic that recognizably identifies its individual or institutional owner, including celebrities who have rights of "publicity." See, e.g., Frampton vs. Frampton Enters, Inc., Case No. D2002-0141 (WIPO, Apr. 17, 2002)(peterframpton.com). [Note: Also, while the Guidebook also addresses rights of an inter-governmental organization ("IGO"), which involves a relatively narrow subset of potential objectors, so it will not be directly addressed here.]
Logistics and Cost-Planning:
Trademark owners and those with prior experience in UDRP procedures will see a familiar face with respect to LRO: the World Intellectual Property Organization ("WIPO"). WIPO has a very informative FAQ page which will cover what would-be objectors need to know (along with model objection and response form), and the ICANN Webinar of Mar. 6, 2013 contains additional information. Suffice to say here simply that with LRO both the filing and expert costs are levied on a flat-fee basis (recall that ICC Expert fees are billed hourly) and parties have the option to choose the size of the panel (either one or three experts). Myself, I always try to garner a three member panel whenever that option is available. All documents are (as with the other three objections) submitted in English and with a limitation of 5000 words (or 20 pages, whichever is less, excluding attachments).
Objection Standards:
Once an objector establishes standing under the relatively straightforward standard above, it must then tackle the merits of the objection. What must the objector show to prevail? Practitioners who are familiar with trademark and domain name disputes will recognize that the LRO standard incorporates elements employed in these areas. A LRO panel must determine whether an applied-for string:
- Takes "unfair" advantage of the distinctive character or reputation of the objector's trademark;
- "Unjustifiably" impairs the distinctive character or reputation of that mark; or
- "Otherwise" creates an "impermissible" likelihood of confusion with the mark.
AGB § 3.5.2. Some of these buzzwords suggest both trademark dilution ("distinctive character" and "reputation") as well as infringement ("likelihood of confusion"). However, rather than delve into all the nuances of the multipart tests used by courts when analyzing these concepts — see, e.g., AMF Inc. v. Sleekcraft Boats, 599 F.2d 341, 348 (9th Cir. 1979) — simply consider this: is it sufficiently likely that a reasonable person will become confused into thinking that a term associates a particular string with one company when it really belongs to another? If not, then look at whether the mark at issue is really well-known as a brand and very unique. If the answer is not an (extremely) emphatic "yes" to either one of these questions, then pursuing an LRO is probably just wasted effort and expense.
So where does one look for help on these issues? Since ICANN has provided a number of helpful guidelines for LRO, we of course will go through those first.
Trademark Factors:
Section 3.5.2 of the Guidebook lays out a number of expressly "non-exclusive" criteria for determining whether an applied-for string creates a likelihood of confusion or will injure the distinctive nature or reputation of a mark. None will strike trademark or UDRP professionals as particularly unusual, although simplicity in concept can belie greater complexity in practice.
1. Similarity in Sight, Sound and Meaning
A LRO panel will examine whether the applied-for gTLD is identical or confusingly similar in appearance, sound or meaning to the objector's mark. For this test, consider the obvious:
Regarding sight, do the mark and the string look visually similar? Since TLDs do not involve graphic representations, colors, fonts and the like, the inquiry essentially comes down to spelling.
As to sound, say the mark and string out loud to yourself. Do they sound essentially the same? See Deutsche Telekom AG v. foxQ, Case No. D2004-0102 (WIPO, Mar. 7 2004) (UDRP complaint denied despite some sound similarity between domains t-online.com and d-online.com).
On meaning: Take into account what term(s) make up the mark and the string. Do they typically have only one connotation or several? This can matter a great deal when examining terms that have generic, descriptive or common dictionary meaning. See, e.g., Advertise.com, Inc. v. AOL Advertising, Inc., 616 F.3d 974, 978-979 (9th Cir. 2010)(advertising.com)(copy of court decision available at: http://1.usa.gov/ZyaxQI); see also Hasbro, Inc. v. Clue Computing, Inc. 66 F. Supp. 2d 117, 133 (D. Mass. 1999) (clue.com).
Simple, right? Well, lawyers have come up over the years with complications even to these apparently straightforward inquiries. Among other things:
- Don't look at two marks together; consumers typically view them separately in the marketplace. Union-Carbide v. Ever-Ready, 531 F.2d 366, 382 (7th Cir. 1975). They also do not recall marks as well as triers of fact who see them continually over days of legal proceedings.
- Consider the trademark and string each as a whole; do not dissect either into components. While courts occasionally give "dominant" components slightly more weight, they do not do so with more generic or descriptive portions of claimed marks — see, e.g., Gateway 2000, Inc. v. Gateway.Com, Inc. 1997 U.S. Dist. LEXIS 2144 (E.D.N.C. 1997) (no likelihood of confusion between "GATEWAY2000" and gateway.com).
- For dilution, a claim may cut a wider swath because the goods or services need not directly compete, but this typically requires greater proof of similarity. See Ringling Bros.-Barnum & Bailey Combined Shows Incorporated v. Utah Div. of Travel Dev.,170 F.3d 449 (4th Cir. 1999) (no dilution between non-competitors' "greatest SHOW on earth" and "greatest SNOW on earth" phrases).
2. Objector's Bona Fide Acquisition and Use of Trademark Rights
A LRO panel also will consider whether the objector legitimately acquired or has used the mark at issue. UDRP colleagues often examine the efficacy of the complainant's asserted rights, including for lack of protectability, lack of authorization or fraudulent procurement.
3. Public Sector Recognition
To what extent does the relevant sector of the public recognize the rightsholder's mark? As in trademark litigation, this may require survey evidence.
4. Applicant's Intent in Applying for the gTLD
This fourth element calls to mind the "bad faith" factor in UDRP and cybersquatting cases. It asks whether the applicant knew of the objector's mark at the time applying for the gTLD, or whether the applicant has a "pattern" acquiring or operating domains confusingly similar to the marks of others. Of course, this applies only to protectable marks, and not to dictionary or generic terms, where intent to confuse either cannot be inferred or is legally meaningless. See, e.g.,Hero, Inc. v. The Heroic Sandwich, Case No. D2008-0779 (WIPO, Aug. 13, 2008) (hero.com).
5. Applicant's Bona Fide Use
This factor allows a panel to evaluate the extent to which the applicant has used (or demonstrably prepared to use) the gTLD corresponding to the objector's mark. While UDRP cases often feature the registration (and little else) of a second-level domain, an applicant for a new gTLD can cite substantial investments in time, resources and money preparing the application and putting backend technology into place. Moreover, using a common word as a domain for its inherent value in attracting internet traffic, even if that term happens to correspond to another's trademark, does not violate the other's rights. Mobile Communication Service Inc. v. WebReg, RN, Case No. D2005-1304 (WIPO, Feb. 24, 2006) (mobilcom.com transfer denied).
6. Applicant's IP Rights
Next, a panel looks at "whether the applicant has marks or other intellectual property rights in the sign corresponding to the gTLD," and, if so, whether such acquisition and use of the sign has been bona fide, and whether the likely use of the gTLD by the applicant is consistent with such acquisition or use." This would pertain to "Dot Brand" TLDs, but not to the many new gTLD applications with generic terms, as these serve no source-identifying function. See Image Online Design v. ICANN, 2013 U.S. Dist. LEXIS 16896, 22-24 (C.D. Cal. 2013) (quoting Advertise.com and finding no protectable interest in plaintiff's alleged .WEB "trademark" — which was also the subject of a prior TLD application) (copy of court decision available at: http://bit.ly/15CWCwV).
7. Applicant Commonly Known by Applied-For TLD
Taking another page from the UDRP playbook, this factor examines whether and to what extent a TLD applicant has been "commonly known by sign corresponding to the TLD and if so, whether any purported or likely use of the gTLD by the applicant is consistent therewith." This will entail case-by-case analysis since, for example, under the UDRP respondents have become "commonly" known by their domain names simply by having formed a company and done business under that name.
8. Likelihood of Confusion
Likelihood of confusion appears both in traditional trademark and domain name disputes, although not in as much detail in the latter setting. Many of the factors above go into a typical judicial likelihood of confusion analysis. Without repeating myself, I simply suggest keeping practical considerations in mind. Among the most important: evidence of actual confusion serves as one of the strongest indicators of a likelihood of confusion, whether from customer service inquiries or other "real-world" data points, or from consumer surveys conducted specifically for the dispute.
9. Other Considerations
I have noted a number of proverbial "bumps" in the objector's "road" above when discussing the Guidebook's specific LRO criteria. However, other overarching considerations also may be back in the mind of a panelist seasoned in trademark and domain name-related matters, and since the LRO factors are explicitly referred to as being "non-exclusive" there would seemingly be no reason not to take them into account.
Fair Use and Free Speech: First, how do free speech-related defenses such as "fair use" or "nominative use" fit into the LRO context? For readers who are not familiar with the concepts, trademark "fair use" typically involves a mark that is capable of describing the goods or services offered under that mark. Although one may own trademark rights in a descriptive term in certain contexts, "such rights will not prevent others from using the word ... in good faith in its descriptive sense, and not as a trademark." Car-Freshner Corp. v. S.C. Johnson & Son, 70 F.3d 267, 269 (2d Cir. 1995). "If any confusion results to the detriment of the markholder, that was a risk entailed in the selection of a mark with descriptive attributes." Id. at 270. Holders of trademark rights in these types of terms would likely face an uphill battle challenging their use in a New gTLD in a generic or descriptive sense.
On the other hand, "nominative" use comes up in situations involving "fair comment" about or criticism of something else. See New Kids on the Block v. New Amer. Pubs., 971 F.2d 302, 308 (9th Cir. 1992). It seems doubtful that an applicant would shell out $185,000 for a new gTLD simply for the narrow purpose of commenting on or referring to something else, making a nominative use defense unlikely to arise.
The Generic Nature of Many TLDs: Similar to the rationale described above involving "fair use" and freedom of speech, would-be objectors should know that, prior to the new gTLD program, courts have held top-level domains as being too generic to even be capable of serving as a trademark in the first place. Cases (in the U.S., at least) uniformly hold that adding a ".com" TLD to an otherwise common word will not confer trademark rights in the combined term. "Because TLDs generally serve no source-indicating function, their addition to an otherwise unregistrable mark typically cannot render it registrable." In Re: Oppendahl & Larson,, 373 F.3d 1171, 1174 (Fed. Cir. 2004) (patents.com). See also Image Online, supra, 2013 U.S. Dist. LEXIS 16896 at 22-24 ("the mark ".WEB" used in relation to Internet registry services is generic and cannot enjoy trademark protection"). Accordingly,expect to see LRO filings being mostly limited to just very unique, highly distinctive "dot-brand" gTLDs and not for generic "dictionary" words.
No Per Se Dilution For Domain Names: Even holders of marks that are considered famous," "well-known" or having a "reputation" may encounter difficulty relying solely upon a dilution claim rather than likelihood of confusion in mounting an LRO challenge. By way of example, courts in the U.S. have steadfastly refused to impose a "per se" (i.e. "blanket") rule for dilution in domain names. "Ownership of a famous mark does not result in automatic entitlement to ownership of the mark as a domain name." Nissan Motor v. Nissan Computer, 2007 U.S. Dist. LEXIS 90487, 45, citing Hasbro, supra, 66 F. Supp. 2d at 133. Trademark-savvy LRO panelists will no doubt take heed of the potential for abuse in dilution claims and instead require the higher likelihood of confusion threshold to be met. See, e.g., Clue.com, supra, at 135, quoting 3 J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition § 24:114 (4th ed. 1996) ("The dilution doctrine in its 'blurring' mode cannot and should not be carried to the extreme of forbidding the use of every trademark on any and all products and services, however remote from the owner's usage").
Conclusion
While LRO would appear to have a somewhat greater likelihood of success than the other three objection types in situations involving particular distinctive and very well-known marks, the same does not hold true for the entire new gTLD landscape, which is populated by a number of applications for generic and descriptive strings,. In the latter scenario, the hill is every bit as tough to climb. Rightsholders would be wise to look carefully for real trademark protection rather than just descriptive uses of their brands in New gTLDs before embarking on the LRO path.
This wraps up my series on new gTLD objections, and I hope that everyone has found them helpful and informative. As always, please feel free to reach out to me at any time with any questions.
I'll see you all in Beijing!
Written by Don Moody, Domain Name & IP attorney in Los Angeles, co-founder of New gTLD Disputes