The Legal Rights Objection (LRO) mechanism administered by the World Intellectual Property Organization (WIPO) on behalf of ICANN is starting to look like the reverse FISA Court of domain name system (DNS) rights protection mechanisms (RPMs) at the top level. In FISA Court, the government always wins – at WIPO, the LRO complainant always loses.
The FISA Court is of course the venue in which the U.S. government requests a surveillance warrant against a suspected foreign intelligence agent and is never denied (well, almost never – since the process started in 1979, 11 requests have been denied out of a total of 33,942 received through the end of 2012[1]). FISA Court proceedings are secret and non-adversarial, with only government lawyers presenting arguments. In contrast, the decisions resulting from LRO proceedings are published, and both complainant and respondent have equal ability to present their arguments. Yet, as of Friday, August 2nd all 28 of the decided LRO complaints, out of a total of 69 filed, resulted in losses for the complainants[2].
The lopsided results arouse suspicions that trademark interests may have been fighting the last war and too intensely focused on potential cybersquatting at the second level (a concern that we believe is overblown due to the low traffic prospects for many new gTLDs). They consequently sought and obtained the Trademark Clearinghouse (TMC), Uniform Rapid Suspension (URS) RPMs – and also proposed other second level remedies such as a Global Protected Marks List (GPML) that ICANN did not grant. Yet they may have failed to adequately consider whether top level protections were sufficient, and in particular whether the threshold standard for successfully prosecuting a LRO was unattainably high.
As Thomas O’Toole observed in an excellent analysis recently posted on Bloomberg BNA’s E-Commerce and Tech Law Blog[3]:
[T]he only surprise the future might hold is the faint prospect that a WIPO panelist might actually rule in favor of a trademark owner…The biggest lesson so far has been that the Legal Rights Objection process is wholly ineffective for trademark owners seeking to knock down proposed domains containing generic strings. In many cases, companies owning marks for terms such as EXPRESS, HOME, VIP, LIMITED, MAIL, TUNES have lost LROs to domain applicants with no intellectual property rights in those terms whatsoever. Pinterest’s failed challenge to Amazon’s proposed .pin domain was particularly ominous for mark owners, because Amazon seems pretty clearly to be moving in on Pinterest’s business…
The reason for this dour outlook can be found in passages from two early LRO rulings: the first one, Right at Home v. Johnson Shareholdings Inc., No. LRO2013-0300 (WIPO, July 3, 2013); and the second, my favorite, Express LLC v. Sea Sunset LLC, No. LRO2013-0022 (WIPO, July 9, 2013)…
In Right at Home, panelist Robert A. Badgley offered the first interpretation of key terms in Section 3.5 of ICANN’s New gTLD Applicant Guidebook. The guidebook uses highly qualified language, directing LRO panelists to decide whether the proposed new domain “takes unfair advantage” of the trademark owner’s rights, or “unjustifiably impairs” the value of the mark, or creates an “impermissible likelihood of confusion” between the mark and the proposed domain.
In Badgley’s view, this language creates a very high burden for trademark-based objections…Obviously, it is going to be very difficult for any trademark owner to demonstrate that a proposed domain is so fishy it satisfies the “something untoward … if not to the level of bad faith” standard…
The second opinion, Express LLC v. Sea Sunset LLC, was one of the better opinions (and I am including the federal court stuff that we wade through every day) I have read in a while. Panelist Frederick M. Abbott carefully summarized the arguments on each side (there are good lessons here for attorneys working on the next round of legal rights objections), and the law that he was required to apply to the dispute. When Abbott turned to the reasoning behind his decision to reject Express LCC’s objection to the proposed .express top-level domain, I got that sense that this panelist was a teeny bit irked that ICANN itself had not made the hard policy choices that the LRO had just dropped in his lap. It’s one thing to ask a panelist to transfer a domain name that might have cost the registrant $10 or so; and it’s quite another to ask a panelist to upset an investment of at least a half-million dollars in a new top-level domain. All based on a trademark registration for a generic term, in a single market, issued by a single government entity. Abbott declined to do it…
These two opinions, taken together, look like a terminal diagnosis for trademark owners with rights in generic terms. Right at Home creates a very permissive standard for what constitutes “unfair advantage” by a domain applicant. Express LLC states, almost categorically, that it is “not reasonable” to allow a trademark owner for a generic term to prevent that term from being used as a top-level domain.
O’Toole goes on to speculate that a trademark owner like Express LLC will have equally poor chances of success if it attempts to block others’ use of such hypothetical second level domains as clothing.express, fashion.express, or shoes.express using the traditional UDRP or the new TMC and URS RPMs — because all of them focus on the domain name to the left of the dot and pay no heed to the gTLD to the right – and because Express holds no trademark rights in those hypothetical generic terms on the left side. We largely agree with his analysis, and suspect that this could be the next big trademark protection issue looming on ICANN’s horizon as brands complain about the necessity and cost of registering relevant domains at a gTLD that matches one of its trademarks but is operated by a third party (and, by the way, it’s not clear that a brand like Express would have any special “sunrise” registration priority for potential second level domains along the lines of O’Toole’s speculation). Other questions arise, such as whether Federal Express should have priority rights to, or a potential infringement claim against, federal.express.
Of course, no one was reckless enough to pay a $185,000 gTLD application fee, plus multiple related legal, consulting, and technical costs, to try to “squat’ at the top level on a unique trademark like Google or Microsoft. But when it comes to a dictionary word like Express, un
less the applicant was so clueless as to propose its use solely for fashionable clothing, the LRO is essentially useless – and that may be the correct result. Indeed, short of halting the program, it’s not really clear how one can have a generic word TLD program without allowing the addition of top level domains that are identical to someone’s trademark for something; and that alone is not sufficient to prove trademark infringement.
Domain industry observers such as Andrew Allemann have opined that “The LRO is working just fine, thank you”[4], weighing in with this analysis:
The LRO was created to assuage fears that someone would cybersquat on a top level domain…Objections were filed against .VIP, .mail, and .home. Now, you tell me: when you hear these terms, what brand do you think is being cybersquatted?
Many of the objections were filed by competing applicants that engaged in trademark frontrunning by obtaining dubious trademarks for the string. Others, such as the United States Postal Service’s objection to .mail, were based on stretched interpretations of a trademark (and that’s being generous).
Another industry analyst, Kevin Murphy, has just declared that he will no longer report automatically on forthcoming WIPO decisions in LRO cases, stating[5]:
The Legal Rights Objection has, I think, said pretty much everything it’s going to say in this new gTLD application round. I’m feeling pretty confident we can predict that all outstanding LROs will fail.
This prediction is based largely on the fact that the 69 LROs filed in this round all pretty much fall into three categories.
- · Front-running. These are the cases where the objector is an applicant that secured a trademark on its chosen gTLD string, usually with the dot, just in order to game the LRO process…
- · Brand v Brand. The objector may or may not be an applicant too, but both it and the respondent both own legit trademarks on the string in question. WIPO’s LRO panelists have made it clear, most recently yesterday in Merck v Merck (pdf) and Merck v Merck (pdf), that having a famous brand does not give you the right to block somebody else from owning a matching famous brand as a gTLD.
- · Generic trademarks. Cases where an owner of a legit brand that matches a dictionary word files an objection against an applicant for the same string that proposes to use it in its generic sense. See Express v Donuts, for example. Panelists have found that unless there’s some nefarious intent by the applicant, the mandatory second-level rights protection mechanisms new gTLD registries must abide by are sufficient to protect trademark rights.
In short, the LRO may be one of many deterrents to top-level cybersquatting, but has proven itself an essentially useless cash sink if you want to prevent the use of a trademark at the top level.
The impact of this, I believe, will be to give new gTLD consultants another excellent reason to push defensive gTLD applications on big brands in future new gTLD rounds.
We largely agree with Murphy’s analysis. There may well be a few cases in the remaining objections to be adjudicated by WIPO in which the complainant prevails. But we suspect that they will be such unique outliers that they will actually reinforce what an exceedingly narrow remedy the LRO has turned out to be. That limited utility has already led one disgruntled complainant/competing applicant to declare, “Seems the entire WIPO LRO process was set-up to fail by ICANN with the guidelines they gave the Panelists.”[6]
And, as Murphy observes, filing a LRO with WIPO is not inexpensive. Total filing fees for a single objection to a single new gTLD application can range from $10,000 for a single-expert panel, and up to $23,000 for a 3-expert panel[7]. On top of the fees there are of course legal costs for the attorney preparing the complaint and related expenses for amassing documentation. If the LRO continues to be a preordained shutout for complainants then it is unlikely that it will be used very much at all in the second round of gTLD applications unless significant – and undoubtedly controversial – changes are made in the adjudication standard to give complainants a higher probability of prevailing.
As for the other objection procedure available to third parties, the Community Objection[8] administered by the International Centre for Expertise of the ICC (International Chamber of Commerce), no judgment can yet be rendered on its efficacy because the Center has yet to issue a single decision on the dozens of cases[9] filed with it. However, with standing limited to established institutions ass
ociated within a clearly defined community objectors first have to surmount that procedural hurdle, and then make the string substantive case that there is substantial opposition to the gTLD application from a significant portion of the community to which the gTLD string may be explicitly or implicitly targeted. Additionally, complaints were voiced at the recent Public Forum in Durban that ICC fees[10] , which begin at 17,000 Euro but can quickly mount given the 450 Euro hourly rate for expert arbitrators, are not affordable for many community organizations.
Summing up, the LRO may well have been an effective deterrent against applications for unique trademarked terms but has so far been useless in regard to applications for generic word trademarks, regardless of the trademark’s strength and legitimacy. Its lack of general applicability also seems to be setting up a series of second level domain disputes that may well be outside the existing scope of the UDRP, URS, TMC, or other available rights protection and prioritization mechanisms. All of this may lead to further debate within ICANN on the appropriate scope of trademark rights protections– as well as litigation being filed by unsuccessful complainants who refuse to take WIPO’s decision as the final judgment on their claims.
The clear lesson to brand owners of generic word trademarks is this: If and when the second round of new gTLD applications commences, if you wish to own your trademark at the top level of the DNS (or at least stop others from owning it) you had better open your wallet and apply for it. That compulsion toward unwanted defensive registrations is almost sure to generate further controversy amid charges that ICANN has developed a Midas Touch-like mechanism for generating perpetual revenues for application and use fees for the protection of dictionary words.
Of course, the inflow of defensive applications for generic word trademarks could be a mere trickle compared to the potential surge of defensive and offensive applications for non-trademarked generic words if ICANN ultimately spurs the Governmental Advisory Committee (GAC) advice to prohibit “closed generic” gTLDs unless they satisfy a high public interest standard. In that scenario, every company of any heft will have to seriously consider gTLD applications for the major language words describing its key activities and products, lest a competitor do so first and lock them away. But that’s an issue falling outside any trademark protection debate and a discussion for another day.