Home / Industry

Why UDRP Panelists Must Follow the Policy: A Look at the Devex.org Decision

Observers of the UDRP are aware that ICANN has substantially abrogated responsibility for oversight of the UDRP. ICANN accredits dispute resolution providers (DRPs) without requiring a contract so that the DRPs are held to no enforceable standards. DRPs, in turn, accredit panelists and similarly hold them to no known standards. Once accredited, panelists can largely do what they please as there is no established mechanism for complaint, investigation, remedial action, or de-accreditation based upon non-observance of the Policy and its accepted interpretations. As a result of the absence of oversight, each UDRP panelist is effectively permitted, without fear of consequence, to fashion his or her own unique interpretation of the UDRP which may be completely at odds with established consensus views. Such errant panelists detract from the credibility of the UDRP and undermine the predictability and stability of the dispute resolution system by issuing decisions which are capricious and unreliable. A clear example of this is the recent decision made by a UDRP panelist in a case involving the domain name devex.org.

On November 27, 2017, the National Arbitration Forum (NAF) issued a decision regarding the domain name, devex.org. The Panelist, who is accredited by both NAF and WIPO, ordered that the domain name be transferred away from its registrant and given to a trademark registrant. The Panelist’s ruling wildly departed from established precedent and amounted to the Panelist’s own personal interpretation of the Policy. The Panelist inexplicably ruled that the disputed domain name was “registered and used in bad faith,” notwithstanding that it was registered long before the complainant’s trademark even existed. By the complainant’s own admission, the complainant did not even adopt DEVEX as a mark until at least 2008, whereas the domain name was registered by the respondent in 2005—three years earlier. The Panelist, therefore, found what amounts to a chronological ‘impossibility’; that the registrant had registered the domain name ‘in order to take unfair advantage of a trademark’ which did not even exist.
It is well-established that it is generally impossible for a domain name to have been registered in bad faith when it pre-dates a complainant’s trademark rights (See, for example, WIPO Consensus View 3.0, at Section 3). Indeed, until recently, the devex.org Panelist himself expressly agreed with this consensus and even went so far as to forcefully expound upon it in an earlier decision from 2010 [1]:

“The undisputed facts in this matter foreclose a finding of bad faith registration and use under Policy 4(a)iii. While Complainant currently has trademark rights in the at-issue mark by virtue of its RIVERON trademark registration, such rights do not magically relate back to the time that Respondent first registered the

domain name, a time well prior to Complainant’s first use of its mark.

Normally speaking, when a domain name is registered before a trademark right is established, the registration of the domain name was not in bad faith because the registrant could not have contemplated the complainant’s non-existent right.” [emphasis added]

It was therefore apparently obvious to the Panelist until relatively recently that it would require “magic”, as he had put it, for a domain name to be registered in bad faith when there was no corresponding trademark right in existence.

However, in the devex.org decision, the Panelist evidently abandoned his previous rational interpretation of the Policy which is widely acknowledged by UDRP panelists as forming the general consensus view [2]. Instead, the Panelist fashioned his own radical and unsupported approach to ‘bad faith registration’. In devex.org, while the Panelist conceded that the established “doctrine recognizes that the registrant/respondent could not have contemplated the complainant’s non-existent rights at the time it registered its domain name and thus could not have acted in bad faith”, he nevertheless claimed that an “exception” exists that entitled him to find bad faith registration even without a trademark in existence.

The Panelist pointed to the WIPO Consensus View 2.0 [3], claiming that it enabled a finding of bad faith registration when a domain name is registered “to take advantage of the confusion between the domain name and any potential [though not yet created] complainant rights”. Indeed there is an exception to the general rule described in the Consensus View 2.0 [4], but as is readily apparent that in order to rely upon the exception, the respondent must be have been “clearly aware of the complainant” and its nascent trademark rights.

Examples of the correct application of the exception are referenced by the Consensus View, and include situations such as: a) where a domain name is registered after a publicized merger between companies but before any new trademark rights in the combined entity have arisen; b) where a domain name is registered as a result of a widely anticipated product launch, but the trademark has not yet been adopted; or c) where the registrant uses inside information about the imminent adoption of a trademark. The key in such exceptional situations is that the registrant is targeting a specific company based on information that a new trademark will shortly be adopted. The exception clearly does not apply where a registrant registers a domain name before the complainant itself is even aware of its mark.

In the devex.org case, there was simply no evidence of any “front running”. There was no evidence of “awareness of the complainant” nor was there any evidence of any “nascent” trademark rights. Rather, the undisputed evidence was that the respondent clearly registered the domain name before it was even a twinkle in the complainant’s eye.

The essence of the Policy is that bad faith requires targeting of a specific trademark both in the registration and use of the disputed domain. As discussed above, such a finding is an impossibility when the trademark does not exist at the time of the domain’s registration. Undeterred, however, the Panelist found that ‘the purpose of the domain name registration was to take advantage of an unidentified trademark owner “at some time in the future” in the event that such a trademark ever came into existence’. The Panelist thereby created out of thin air a brand new concept of ‘anticipatory’ bad faith registration which is totally unsupported by the Policy and consensus views on its interpretation.

According to this Panelist, an innocent person who registers a domain name free of any trademark rights whatsoever by anyone, anywhere, which 13 years later becomes attractive to a company who has on its own volition, decided to launch a product under a corresponding brand even though the domain name is already spoken for, is guilty of “anticipatory” bad faith. In other words, an undeniably good faith registration can be transformed into a bad faith registration simply because the registrant speculatively registered the domain name in the hopes or expectation that one day someone would want to adopt a corresponding brand and would, therefore, be willing to purchase the rights to the corresponding domain name.

There is, however, no such thing as “anticipatory bad faith” in the Policy, nor is any such notion an accepted interpretation of the Policy. In fact, the opposite is true; speculating in un-trademarked, brand-able domain names is well established under the Policy as a “legitimate interest”. For example, as held in Alphalogix, Inc. v. DNS Services d/b/a MarketPoints.com – New Media Branding Svcs., NAF Claim Number: FA0506000491557, and as unanimously followed by the three-member Panel in Arrigo Enterprises, Inc. v. PortMedia Domains, NAF Claim Number: FA1304001493536, the business of creating and supplying names for new entities is a “legitimate activity in which there are numerous suppliers in the United States”. Indeed, there is a lawful massive, multi-million dollar aftermarket for the trading in un-trademarked, descriptive, or generic domain names which may be attractive to end-users, as demonstrated by, for example, Sedo and Godaddy, both of which operate successful domain name marketplaces. Under the Panelist’s view, any registration in the hopes of it becoming valuable to a company who decides to adopt a corresponding brand would render the registration as having been undertaken in “bad faith” and would effectively outlaw domain name speculation, period.

Moreover, the Panelist took the essence of the Policy, namely intent, and flipped it on its head. The Panelist expressly conceded that “[2016] was arguably the first time when Complainant had a claim against Respondent”, thereby implicitly admitting that for a period of 13 years, from the date of registration in 2003 onwards to 2016, the registration was in good faith. Nevertheless, the Panelist found the registration in ‘bad faith’ from the very moment that the registration was made. It was not until the moment that the complainant chose to adopt as a trademark a term which it likely knew had already been registered as a domain name by another party, that the initial registration 13 years prior supposedly “revealed” itself to be a ‘bad faith registration’. The absurdity of the Panelist’s position is that the only “intent” exercised at that moment, was by the complainant in adopting the trademark, not by the registrant who had registered the domain name many years before. Accordingly, it was, in fact, the complainant itself who instigated and indeed created this supposed bad faith.

So why would the Panelist, who previously subscribed and indeed forcefully supported the consensus view of bad faith registration under the Policy, purport to create his own radical interpretation? What apparently troubled the Panelist was that “the logic underpinning the [general] rule d[id] not square with the instant circumstances”, i.e. he was frustrated that the Policy does not permit a remedy when only bad faith use exists, but not bad faith registration. He had found that the disputed domain name had been misused but couldn’t find any bad faith registration under the accepted view, so he unilaterally changed the definition of ‘bad faith registration’ in order to support his desire to transfer the domain name. Instead of acknowledging that the UDRP has its limits, he purported to expand those limits on his own.

Regardless of the Panelist’s motivations, what he did was wrong for five fundamental reasons:

First, the Policy was not intended nor designed to be used as an all-encompassing form of trademark protection for every aggrieved trademark owner. The limited scope of the UDRP was discussed by the ICANN staff in its Second Staff Report on Implementation Documents for the UDRP, dated, October 24, 1999, at paragraph 4.1(c). The Report expressly stated that the Policy was intended to apply to “only a small, special class of disputes”, not every kind of trademark dispute that involves a domain name. Trademark infringement disputes involving domain names without bad faith registration are left to the courts to decide. Moreover, The Final Report of the WIPO Internet Domain Name Process of April 30, 1999, indicates that the UDRP must be limited in scope “to cases of bad faith, abusive registration of domain names that violate trademark rights”—which implicitly refers to trademark rights which actually exist at the time of domain name registration—not trademark rights which may one day in the future arise. Accordingly, the Panelist improperly invoked the UDRP to apply to a class of case which it clearly wasn’t intended to address and indeed which the Policy purposefully excluded.

Second, the Panelist improperly took it upon himself to rewrite the Policy. It is not for panelists to create policy. The UDRP was crafted nearly 20 years ago and is the product of careful balancing between competing rights and interests. It was drafted in the manner which it was for good reason and with input from numerous stakeholders after careful study. If a particular panelist wishes that he or she could provide a remedy which the Policy does not permit, the panelist must nevertheless acknowledge and adhere to the Policy’s limits, which are “a feature” and “not a flaw”, as expressed in the Second Staff Report [5]. The place for review and any possible reform of the Policy is within an established policy development framework, namely the ICANN Working Group mandated to review Rights Protection Mechanisms [6] (the “RPM Working Group”). The RPM Working Group is where deliberations amongst stakeholders occur, and if changes are to be made to the Policy taking into account all considerations and interests, it is there and not on the initiative of a particular Panelist.

Third, the Panelist improperly sought to conflate bad faith registration with bad faith use. Under the Panelist’s misguided approach, ‘bad faith registration’ would become indistinguishable from bad faith use, since, upon bad faith use, bad faith registration would be imputed to have existed at the time of registration. It is enshrined in the Policy itself that in order to meet the third part of the three-part test under the UDRP both bad faith registration and bad faith use must have occurred, [7] and these are distinct concepts which must be respected and abided by.

Fourth, the Panelist improperly sought to effectively re-introduce the equivalent of ‘retroactive bad faith registration’ [8], which is an approach that has been unequivocally discredited. [9] The outlier and now discredited theory of so-called retroactive bad faith posited that a good faith registration could be converted into a bad faith registration as a result of the registrant subsequent use of the domain name. That is essentially the same approach in substance as espoused by the Panelist since the Panelist imputed bad faith registration to the respondent as a result of his subsequent use of the domain name. Accordingly, whether framed as ‘retroactive bad faith’ or as ‘anticipatory bad faith’ as the Panelist did, both subscribe to the same errant view which has been unequivocally rejected and discredited by the preponderance of UDRP panelists.

Fifth, the Panelist improperly sought to effectively prohibit lawful business practices, which is a decision that is far beyond the ambit of the UDRP, let alone a particular Panelist. Domain names are of business interest to more than trademark owners, who are just one of many stakeholders in the Internet community. Domain name investors who invest considerable sums and expend considerable resources in acquiring and managing valuable domain names have an important interest in domain names as well. For more than 20 years, domain name investors have lawfully complied with the Policy and national laws and have built their respective businesses on the expectation and assurance that they would not be deprived of their property and livelihoods. By fabricating his own radical view which effectively renders it unlawful for a domain name investor to speculate on increasing value of a particular domain name, the Panelist wrongly sought to destroy a vibrant, important, respected industry which had heretofore been considered entirely lawful, merely so that he could provide his own remedy in circumstances which did not permit a remedy under the UDRP.

The devex.org case provides an important lesson in the danger of a lack of ICANN oversight over the UDRP process. Left to their own devices, some panelists will attempt to create their own version of the Policy which is at odds with established interpretive norms and consensus views. Moreover, the dispute resolution providers must also bear responsibility for those whom they accredit as panelists and assign to decide disputes when these panelists undermine the very Policy that the dispute resolution provider is entrusted with implementing. The dispute resolution provider is currently the only body empowered to discipline or de-accredit errant panelists. All panelists must elevate scrupulous adherence to the Policy and consensus views to the utmost of importance, even if they happen to personally feel that they can come up with a better solution themselves. Otherwise, the UDRP’s credibility as a trustworthy and relative predictable dispute resolution mechanism will suffer to the detriment of all stakeholders.

The UDRP suffers from the recurring problem of panelists who take it upon themselves to rewrite the Policy. These panelists are predominantly drawn from the ranks of lawyers whose background is in aggressively enforcing alleged trademark rights on behalf of their corporate clients. A panelist who wishes to expand the scope of the Policy to make it a more effective tool for policing asserted trademark rights will find many natural allies among other likeminded panelists. For instance, in the Camilla.com decision from 2015, a group of three panelists, all with backgrounds as IP litigators, collaborated in rewriting the Policy to invent an ongoing duty imposed on domain registrants to monitor new trademark uses worldwide, and an obligation to steer clear of any new trademarked use wherever it may arise. The ICA issued a statement at the time decrying that the panelists in the Camilla.com dispute would so utterly ignore their duty to be faithful to the Policy as written. [10]

The ‘Retroactive Bad Faith’ theory that the panelists in Camilla.com employed in justifying their transfer order has since been “utterly discredited” by the panelist community, and has been explicitly rejected in the latest version of WIPO’s Overview, as the Panelist in devex.org is likely well aware [11]. The rationale in devex.org is yet one more unfortunate variation on the discredited ‘Retroactive Bad Faith’ theory which the Panelist in devex.org is trying to resurrect with apparently little care for the necessity for a clear and consistent application of the UDRP.

While the community of panelists admirably attempted to police themselves by rejecting the RBF theory in the WIPO Overview 3.0, clearly that is not sufficient. A system developed to benefit trademark interests, operated by groups created to represent trademark interests, conducted in a forum selected by the trademark owner, and where the panelists are largely selected from the trademark bar and given free rein to interpret the Policy as they desire, is not a system designed to produce fair or consistent outcomes. Greater community oversight is needed, and the significant bias built into the current system needs to be removed so that the interests of trademark holders and domain registrants are more fairly balanced.

[1] See, Riveron Consulting, L.P. vs. Stanley Pace, NAF Claim No. FA1002001309793 (April 12, 2010)

[2] See, WIPO Consensus View 3.0, at Section 3.

[3] Version 3.0, in effect at the time of the devex.org decision, superseded Version 2.0 referenced by the Panelist in the devex.org decision.

[4] And also in the 3.0 version.

[5] See, Paragraph 4.1(c) therein.

[6] See, Review of all Rights Protection Mechanisms (RPMs) in all gTLDs PDP Working Group Home.

[7] See, Duction, Inc. v. John Zuccarini, d/b/a The Cupcake Party & Cupcake Movies, WIPO Case No. D2000-1369)

[8] See, Internet Commerce Association, The Rise and Fall of the UDRP Theory of Retroactive Bad Faith, CircleID, March 8, 2017.

[9] See, Etsy, Inc. v. Domain Administrator / Portmedia Holdings Ltd., Claim Number: FA170001753224 (November 27, 2017) (“Complainant’s arguments find no support within any exception of the now largely-discredited Mummy/Octogen line of decisions”); and also see WIPO Consensus View 3.0 at Paragraph 3.2.1, “this so-called retroactive bad faith registration [approach] has not been followed in subsequent cases”.

[10] See, “Initial Statement of the Internet Commerce Association Regarding the Camilla.com UDRP Decision”, December 14, 2015.

[11] See Note 9, supra.

By Internet Commerce Association, Non-Profit Trade Organization Representing Domain Name Owners

Founded in 2006, the Internet Commerce Association (ICA) is a non-profit trade organization representing domain name investors, website developers and related companies.

Visit Page

Filed Under


The correct, sensible approach Nat Cohen  –  Dec 14, 2017 6:55 PM

In the DLL.com decision issued on December 7th, panelist Nick Gardner found-

“If the Respondent had no knowledge of the Complainant when registering the Disputed Domain Name then bad faith cannot be established.”


That is a sensible, common-sense threshold question for whether the possibility of a bad faith registration exists.  If the Panelist in devex.org had followed this common-sense approach, the travesty of the misguided devex.org decision could have been avoided.

Excellently put, but not sure what kind Gerald M. Levine  –  Dec 14, 2017 8:58 PM

Excellently put, but not sure what kind of “oversight” can reasonably be expected from ICANN or the providers. There are certainly (rogue) panelists who go their own way, but the consensus has been remarkably steady on the jurisprudence in balancing parties’ rights (as Nat indicates in citing DLL). The Camilla ACPA case ended with a stipulation to vacate the UDRP award. Trial level judges also make errors of law, but they (the errors) can be corrected on appeal. Question: Should the UDRP be amended to include a provision for an administrative appeal (that does not prejudice a respondent’s right under 4k if that review is affirmed)?

Oversight Nat Cohen  –  Dec 14, 2017 9:24 PM


An administrative appeal would be one form of oversight.  Another form of oversight would be if the provider de-accredited a panelist, such as in the devex.org case, who demonstrates such disregard for the Policy.  By continuing the accredit the panelist, and by continuing to assign cases to such a panelist, the provider becomes complicit in the undermining of the Policy it has committed to enforce.



De-accreditation of a panelist seems to be Samit Madan  –  Dec 15, 2017 7:29 AM

De-accreditation of a panelist seems to be a sensible approach, would provide for some pause to rogue panelists who decide to rewrite policy.

Zak, Thanx for a great article about Louise Timmons  –  Dec 15, 2017 9:48 PM

Zak, Thanx for a great article about the danger of leaving panelists to their own devices. Is there not grounds to sue ICANN to change its language describing providers from, “approved?” Here is the definition of, “approve:”

officially agree to or accept as satisfactory

- Dictionary.com

With limited checks in place for UDRP and URS providers, ICANN misrepresents its service:

List of Approved Dispute Resolution Service Providers

bold mine
- https://www.icann.org/resources/pages/providers-6d-2012-02-25-en

Where is the protection for the Registrant in the following provision outlined by ICANN:

In recognition of that potential [of UDRP provider abuse], ICANN commits that substantiated reports of UDRP provider non-compliance with the UDRP or the Rules will be investigated. If the investigations uncover issues of UDRP provider non-compliance, ICANN will work with the affected UDRP provider to determine if the issue can be remedied. If the issue cannot be remedied, and the UDRP provider cannot – or refuses – to return to acting in conformity with the UDRP, ICANN will take action, which might include revocation of its approval of the UDRP provider, taking into account issues relating to the transferring or completion of pending matters before that provider.

bolds mine
The report against a UDRP providers has to be SUBSTANTIATED, first.
IF ICANN finds an issue of non-compliance, it WORKS WITH THE PROVIDER to find out if the “issue can be remedied.” What does that mean? The provider acknowledges his mistake and goes on? By that time, a business using a domain can be affected, a sale can be lost, or the domain could be transferred to a foreign Registrar like eName, which traffics stolen domains, and which ICANN< Verisign, and the major Registrars faciitate its trafficking . . .

Here is the reference for the above Louise Timmons  –  Dec 15, 2017 9:52 PM

Here is the reference for the above quote:

UDRP Providers and Uniformity of Process – Status Report

Commenting is not available in this channel entry.
CircleID Newsletter The Weekly Wrap

More and more professionals are choosing to publish critical posts on CircleID from all corners of the Internet industry. If you find it hard to keep up daily, consider subscribing to our weekly digest. We will provide you a convenient summary report once a week sent directly to your inbox. It's a quick and easy read.

I make a point of reading CircleID. There is no getting around the utility of knowing what thoughtful people are thinking and saying about our industry.

Co-designer of the TCP/IP Protocols & the Architecture of the Internet




Sponsored byDNIB.com

Domain Names

Sponsored byVerisign

Brand Protection

Sponsored byCSC


Sponsored byVerisign

Threat Intelligence

Sponsored byWhoisXML API

IPv4 Markets

Sponsored byIPv4.Global

New TLDs

Sponsored byRadix