The modus operandi of the Internet Corporation for Assigned Names and Numbers (ICANN) is achieving consensus. This also holds true for the principal rights protection mechanism that emerged from a two-year round of debates organized by the World Intellectual Property Organization (WIPO) that ICANN implemented in 1999 as the Uniform Domain Name Dispute Resolution Policy (UDRP). Consensus rules; not precedent, although consensus inevitably becomes that. The concept of consensus was highlighted in the WIPO Overview of WIPO Panel Views on Selected UDRP Questions, Original Edition (2005): “On most of these issues, consensus or clear majority views have developed.” It also pointed out that for certain other questions there had emerged (not surprisingly given the diversity of panelists from different jurisdictions) different views. The word “consensus” appeared 21 times in the Original Overview.

In the Second Edition of the Overview (2011), “consensus” appears 28 times: “This WIPO Overview was originally created and has been updated and expanded in recognition of the need that has been expressed to identify, as much as possible, consensus among UDRP decisions, so as to maximize the consistency of the UDRP system.” It continued the practice of the First Edition by identifying where there was divergence, “View 1” and “View 2”. The Third Edition (2017) eliminates the word “consensus” as it applies to decisions for the simple reason that there is now more consensus and that for the most part differences have converged to become “precedent” even though only informally acknowledged:

While the UDRP does not operate on a strict doctrine of binding precedent, it is considered important for the overall credibility of the UDRP system that parties can reasonably anticipate the result of their case.

I say “for the most part” because while the diversity of view has been narrowed by a jurisprudence endorsed by consensus, some panelists continue to act as though there was no consensus which has a negative effect on “the overall credibility of the UDRP system.” A recent example of a nonconformist decision is the <devex.org> case, Developmentex.com, Incorporated v. Manuel Schraner, FA171000 1755537 (Forum November 27, 2017). This decision has been treated to a blistering attack from the Internet Consumer Association here in CircleID. (I also discussed the decision earlier in a separate essay). When the agreed upon consensus is willfully upset as it is in “devex,” it introduces an unacceptable bias into the ICANN process. (Appointed Panels are supposed to be neutral). One of the comments to the ICA essay is that de-accrediting panelists “seems to be a sensible approach [in that it] would provide for some pause to rogue panelists who decide to rewrite policy.” I will return to this de-accrediting suggestion in my conclusion, where I will also point out a couple of other possibilities).

A general view formed early in the jurisprudence is that the UDRP should not be a roulette wheel. It “should consist of more than, ‘[i]t depends [on] what panelist you draw.’” Time Inc. v. Chip Cooper, D2000-1342 (WIPO February 13, 2001) (<lifemagazine.com>). The goal (as stated in many UDRP decisions) is achieved through “a strong body of precedent” which “is strongly persuasive” even if not binding. Pantaloon Retail India Limited v. RareNames, WebReg, D2010-0587 (WIPO June 21, 2010) (“Whether [the consensus in holding that a respondent in the domain name business] is justified may be a matter for debate, but in the opinion of the Panel there is a strong body of precedent which, though not binding, is strongly persuasive.”)

Consistency with consensus views (the “strong body of precedent”) is officially encouraged by WIPO, and presumably also by the other providers. Although the devex case comes from the Forum, there are some past WIPO decisions in which the reasoning and conclusions are clearly in opposition to the law. A WIPO email I received on December 20 reported that on “October 23 some 100 WIPO Panelists met at WIPO Headquarters in Geneva to discuss recent jurisprudential developments in cases administered by the WIPO Center.” The email emphasized that “the meeting provided a venue for presentations and discussions to help maintain jurisprudential consistency” (My emphasis).

Panelists whose rulings are inconsistent with the jurisprudence mainly favor trademark owners whose rights did not exist when the domain names were registered. If the jurisprudence is properly applied, there can be no cybersquatting claim under the UDRP (or for that matter the Anticybersquatting Consumer Protection Act). Priority is one of the Policy’s basic principles. De Lage Landen International B.V. v. Steve Thomas, D2017-2045 (WIPO December 7, 2017) (<dll.com>) and Virgin Enterprises Limited v. Domain Admin/This Domain is for Sale, Hugedomains.com., D2017-1961 (WIPO December 11, 2017) are the antithesis of Developmentex.com: “Complainant’s registered rights in the letters ‘DLL’ as a word mark dating from 2007, long after the Disputed Domain Name was registered.” In Virgin Enterprises registering “domain names that include generic words for the purposes of selling them ... can be legitimate and are not in themselves a breach of the Policy, so long as they do not encroach on third parties’ trademark rights.” In both cases, the domain names were registered before the existence of a mark. Potent mark as VIRGIN is, Complainant does not own the word; its exclusive right does not extend to all phrases in which virgin is combined with another word.

In the devex case the Panel willfully misstates the law when it intones that “[t]he fact that Complainant had no rights in DEVEX at the time Respondent registered <devex.org> does not permit Respondent to exploit another’s after-acquired rights.” Rights acquired after registration of the domain name are not actionable under the UDRP, period.

That such maverick decision-making comes about is due in large measure to the repudiation of consensus in two 2009 sharply reasoned decisions from (surprisingly!) the Panel responsible for the binary view becoming the bedrock principle of the UDRP. When in the first UDRP decision (2000) the Panel announced that “[i]t is clear from the legislative history that ICANN intended that the complainant must establish not only bad faith registration but also bad faith use” he created precedent by interpreting what he believed was the intention agreed upon by the multiple constituencies engaged in birthing the UDRP, World Wrestling Federation Entertainment, Inc. v. Michael Bosman, D1999-0001 (WIPO January 14, 2000). However, in Octogen Pharmacal Company, Inc. v. Domains By Proxy, Inc. / Rich Sanders and Octogen e-Solutions, D2009-0786 (WIPO August 19, 2009) the first Panel took back his interpretation of the UDRP by formulating the so-called “retroactive bad faith” principle: “[I]n this Panel’s view bad faith registration can be deemed to have occurred even without regard to the state of mind of the registrant at the time of registration, if the domain name is subsequently used to trade on the goodwill of the mark holder.” In Guru Denim Inc. v. Ibrahim Ali Ibrahim abu-Harb, D2013-1324 (WIPO September 27, 2013) (<truereligion.com>) the same Panel (in dissent) explained that “[i]t would be much easier for this panelist to maintain that his original decision [approving the binary concept] was correct, and not recant. But in view of the evidence [of the correctness of the unitary view], I am unable to do so.” The majority got it right by following consensus:

In the present case, when the Respondent registered the disputed domain name, the trademark did not exist. Respondent was therefore entitled to continue using the domain name <truereligion.com> as he had been using it, or to use it for any purpose whatsoever, so long as he did not intentionally use it to profit from the goodwill associated with Complainant’s later created trademark.

Nevertheless, several panelists agreed with the Octogen e-Solutions analysis (and some continue to agree, as we see in the devex case). In Big 5 Corp. v. EyeAim.com / Roy Fang, FA130800 1513704 (Forum October 11, 2013) (with dissent in favor of consensus. <big5.com>) the majority (hewing to the new construction) held that it “deems Respondent’s 2012 renewal of the disputed domain name to be the date on which to measure whether the disputed domain name was registered and used in bad faith for purposes of paragraph 4(a)(iii).” Similarly in Milly LLC v. Domain Admin, Mrs. Jello, LLC, D2014-0377 (WIPO May 25, 2014) (<milly.com>) in which Complainant requested the Panel consider the retroactive bad faith or the unified concept approach which the Panel obligingly did: “[t]he fact Respondent may have registered the Disputed Domain Name prior to Complainant’s acquisition of trademark rights does not per se preclude a finding of bad faith under the circumstances of this case for the purposes of paragraph 4(a)(iii).” Ditto Jappy GmbH v. Satoshi Shimoshita, D2010-1001 (WIPO September 28, 2010) and other cases, all of which were enabled by the Octogen e-Solutions decision and all contrary to precedent which holds that paragraph 4(a)(iii) is a unitary concept.

Interestingly, the Octogen e-Solutions Panel appears more recently to have realigned with the consensus view. In Group One Holdings Pte Ltd v. Steven Hafto, D2017-0183 (WIPO March 28, 2017) a three-member panel including the first panel unanimously rejected the retroactive bad faith reasoning thereby essentially declaring it a dead end. The interpretation that the UDRP demands proof of bad faith registration AND (not “OR”) bad faith use quickly became the consensus view, and still is (the devex Panel’s misapplication of the law is the equivalent to putting a finger on the scale in favor of a party having no actionable claim either under the UDRP or the ACPA. The latest WIPO Overview even repudiates that reasoning).

Since the only legal challenge of erroneous awards (in the U.S. at least) is a lawsuit under the Anticybersquatting Consumer Protection Act (ACPA) few domain name losers have the stomach to afford or endure an expensive lawsuit. But several have. For example, in Camilla Australia Pty Ltd v. Domain Admin, Mrs Jello, LLC., D2015-1593 (WIPO November 30, 2015) the Respondent, a domain investor, challenged the award under the, Mrs. Jello, LLC v. Camilla Australia Pty Ltd. 15-cv-08753 (D. NJ 8/1/2016) and it was vacated by stipulation, a result that should (if Panels like the one in the devex case are paying attention) should clarify the law as it ought to be applied in UDRP disputes. (Telepathy, Inc. has also gone the ACPA route to vindicate its rights with similar results including recovery of attorney’s fees). (But, if the facts are not right, investors can also get hammered in ACPA cases. They too should have sought professional advise).

So what can be done about panelists who willfully apply their own law? Is de-accreditation of panelists the right answer? Certainly, panelists should be educated in the jurisprudence; they should know what the law is and apply it consistently. It is a disgrace when they are clearly in error; when they apply wrong law! But, I do not believe there would be consensus for de-accreditation and it certainly would not be approved by providers. Another solution for which there may be consensus is creating an appeals procedure similar to Nominet UK’s and a differently phrased version of a rule ICANN introduced in 2013 for the Uniform Rapid Suspension System (URS). One other possibility would be the formation of a corporate entity funded by the domain community to retain counsel to represent investors in challenging decisions inconsistent with UDRP jurisprudence, repaid from recovery under the ACPA of attorney’s fees and damages.

Final thought: the reason there is a secondary market in domain names at all (it seems to me) is due to the first Panel’s original (and luminously reasoned) construction in World Wrestling. Had he construed the Policy in 2000 as he later did in Octogen there would never have developed a secondary market and we would not be discussing what to do with panelists who rule in favor of complainants whose marks postdate domain name registrations. (In other words, the devex ruling would have emerged as consensus!)