The general run of Uniform Domain Name Resolution Policy (UDRP) decisions are unremarkable. At their least, they are primarily instructive in establishing the metes and bounds of lawful registration of domain names. A few decisions stand out for their acuity of reasoning and a few others for their lack of it. The latest candidate of the latter class is NSK LTD. v. Li shuo, FA170100 1712449 (Forum February 16, 2017) (<skfnsk.com>) (I’ll call this the “Li shuo dispute” to distinguish it from other NSK cases discussed further below). It is an example of inconsistency in the application of law. In this case, the Panel denied the complaint. Inconsistency this time affects the trademark owner; other times it has affected the domain name registrant.

Inconsistency in the application of law is always a problem because it affects the predictability of result. With judicial decisions, there’s always higher authority. In addressing the issue in UDRP proceedings, there are three forms of inconsistency: (1) diversity of view; (2) different assessments of facts and applicable law; and (3) “manifest disregard of the law.” Of the first, the World Intellectual Property Organization (wearing its provider hat) acknowledges diversity of view as a feature of the process: Overview 2.0 preliminary discussion reads “[o]n most of these issues, consensus or clear majority views have developed. Certain other questions attract a diversity of views.” While there is not a strict equation between “diversity of view” and the two other forms of inconsistency all three affect predictability, but the last is the most pernicious, which is where the Li shuo dispute belongs.

Not surprisingly for a regime that has no review mechanism, panelists must find their own way which mostly means following consensus or precedent and some go their own way. WIPO Overview of WIPO Panel Views on Selected UDRP Questions, Second Edition (“WIPO Overview 2.0”) states that “[t]he UDRP does not operate on a strict doctrine of precedent” but then hedges by admitting that “panels [nevertheless] consider it desirable that their decisions are consistent with prior panel decisions dealing with similar fact situations.” Paragraph 4.1. Whether we call it consensus or precedent, the agreed upon aim is predictability which is achieved through consistency. “Consistency” ( 4.1 continues) “ensures that the UDRP system operates in a fair, effective and predictable manner for all parties, while responding to the continuing evolution of the domain name system.”

Desirable though consistency is, however, panelists are not always in agreement on contested issues; this is the diversity of view construct. For example, the Overview records two views in assessing fair use. (Overview 1.0 superseded by 2.0 also flagged “majority” and “minority” views, no longer identified in 2.0). But these “diversities of view” don’t generally express themselves as differences between WIPO and Forum appointees. In other words, there isn’t a WIPO view and a Forum view.

While there is not a strict equation between “diversity of view” and the two inconsistencies mentioned above they all affect predictability. Who the provider appoints to arbitrate the dispute could make the difference to its outcome. Doug Isenberg addresses this issue in his essay “When Two Trademarks Aren’t Confusingly Similar to One Trademark,” circleid essay (February 23,2017). The domain name in the Li shuo dispute combines two marks owned by unrelated entities. In denying the complaint, the Panel explained that Complainant lacked standing and gave as its reason that Complainant “alleges no nexus between it and the owner of the [SKF] mark. As such, Complainant essentially has standing to bring this claim regarding the NSK mark but not the SKF mark.” (Note the emphasis on “nexus” which I’ll explain briefly in conclusion). Mr. Isenberg cautions brand owners to brush up their knowledge of the UDRP and be prepared for the contingency of panels going their own way—“As in all domain name disputes, nothing should be taken for granted.”

The NSK decision was also noticed by Evan Brown in his blog udrptracker: “[t]he case” (Mr. Brown says) “serves as an example of a panel departing from the ordinary determination that a disputed domain name incorporating the complainant’s mark as a whole will suffice to demonstrate confusing similarity.” The Li shuo decision is indeed a departure from expectations, and the law! Mr. Brown doesn’t say so directly, but I think he thinks the decision is a headscratcher and not consistent with either “consensus” or “precedent.” After all, the domain name does incorporate the whole of Complainant’s mark. Li shuo is not the first dispute involving combined trademarks yet to read the decision one can be excused for thinking it is.

As evidence that the Li shuo decision is a departure Mr. Isenberg aptly notes several other Forum decisions on combined marks that went the other way even though the second mark owner was absent from the proceedings. NSK LTD. v. niu yanqing / Beijing Jia virtue International Trade Co. Ltd., FA161000 1697729 (Forum November 14, 2016) (<skf-nsk-bearings.com>); NSK LTD. v. DINA MCINTIRE, FA1701001712452 (Forum February 21, 2017) (nsk-fag-skf-ntn.com>). Respondents also defaulted in these decisions. So Li shuo is an unfortunate inconsistency of the third kind.

The Li shuo Panel’s theory appears to have first been offered in NIKE, Inc. and Nike Innovate, C.V. v. Mattia Lumini and Yykk Snc. FA1606001679233 (Forum July 15, 2016) (<nikegoogle.com>) another “outlier” (as Mr. Isenberg aptly notes). The theory comes from nowhere; is in fact fabricated from whole cloth. Neither the Nike Panel nor the Li shuo Panel cite any authority for denying the complaints. In the Nike case, Complainant wasn’t satisfied so it refiled with Google as a joint Complainant and the domain name was transferred. NIKE, Inc., Nike Innovate, C.V., and Google Inc. v. Mattia Lumini / Yykk Snc, FA160800 1687597 (Forum September 20, 2016). This is the model for NSK to follow, assuming it can persuade SKF to join in the proceedings.

Applying a new theory without any underpinning of reasoning or authority undercuts predictability in such a random way that there’s no way to prepare other than Mr. Isenberg’s suggestion with combined marks to have both mark owners join forces. WIPO Panels have explored this issue of combined marks incorporated into domain names more deeply and thoroughly than Forum’s.

In contrast to the NSK and Nike Panels, the Panel in Incase Designs Corp. v. Rogenie LLC, Rogenie Cordero, D2012-1491 (WIPO September 12, 2012) (<iponeincase.com> reasoned its way to a conclusion that preserves predictability: “[t]here is no indication in the record that Apple Inc. would consent to the remedy of transfer to Complainant, and indeed, Complainant appears to be disinclined to request Apple Inc.‘s authorization. In these circumstances, therefore, the Panel has determined that it is most consistent with the Policy and its injunction of efficiency not to delay the proceedings by ordering Complainant to seek such authorization.” But in this case, the Panel granted cancellation not transfer:

Complainant initially requested transfer of the disputed domain name. Later, though, Complainant expressed concern that Apple Inc. would not quickly consent to transfer the disputed domain name to Complainant. Therefore in its amended Complaint, while still requesting transfer, Complainant added a request in the alternative for the remedy of cancellation.

There’s precedent for this caution by Complainant’s counsel and Panel, which I’ll get to in a moment.

The same Panel returned to the issue three years later in Kabbage, Inc. v. Oneandone Private Registration 1, 1&1 Internet Inc.—www.1.and1.com/oert Hanssen, Ridiculous File Sharing,, D2015-1507 (WIPO November 20, 2015) (<kabbage4amazon.com>). Citing earlier authority, it held that “where Complainant has fulfilled all the elements of Policy paragraph 4(a), expressed a strong preference for transfer over cancellation, and demonstrated to the satisfaction of the Panel that it is cognizant of its obligations to respect the rights of third-party trademark holders, the Panel is willing” to grant the request to transfer. The Panel cited a case from 2006 to support the transfer. “As others have written,”

neither the Policy nor the Rules expressly require the consent of a third party and previous panels have accepted complaints request[ing] that a domain name may be transferred to the complainant, noting that such decision would be expressly without prejudice to any rights, which may be asserted by third party trademark holder. Decathlon SAS v. Nadia Michalski, supra. See, e.g., F. Hoffmann-La Roche AG v. Bob, D2006-0751 (WIPO July 20, 2016) (<viagra-xenical-pharmacy.com>).

There is a legitimate and understandable question as to whether the proper remedy in combined marks disputes is cancellation or transfer. In Incase Designs the Panel felt that “cancellation [but not transfer] may be an apt remedy when there is no consent by the third-party trademark holder.” Other panelists are of the view that while “a complainant ideally should try to obtain the third party’s consent before seeking transfer” it is not absolutely necessary. Dr. Ing. h.c. F. Porsche AG v. Automotive Parts Solutions, D2003-0725 (WIPO November 17, 2003) (<911hyundai.com>), cited in Lilly ICOS LLC v. Tudor Burden, Burden Marketing, D2004-0794 (WIPO December 20, 2004).

Several UDRP panels have reached the opposite result, granting transfer in the absence of third-party consent but in doing so emphasizing that “such decision would be expressly without prejudice to any rights, which may be asserted by third party trademark holder,” citing the Decathlon and F. Hoffmann-La Roche decisions.

There is not time to explore fully the (unconscious) theory underlying the Nike and Li shuo decisions but I suspect it may come from the Forum’s Supplemental Rules on multiple complainants, namely they have to demonstrate a “nexus” which neither Complainant in the two cases could do—“multiple persons or entities who have a sufficient nexus who can each claim to have rights to all domain names listed in the Complaint” (Rule 1(e). Apparently in the reasoning of these two panelists combined (or multiple) marks have become equivalent to multiple complainants. This is a bankrupt theory; it threatens the integrity of the UDRP. If predictability is an issue which it obviously is for both trademark owners and domain name registrants, then these decisions should be scratched as dead ends and only referred to as examples of unacceptable inconsistencies.