Trademark owners in the U.S. have a choice in suing for alleged cybersquatting: either the Uniform Domain Name Dispute Resolution Policy (UDRP) or the Anticybersquatting Consumer Protection Act (ACPA). Of the two, the UDRP is far and away the forum of choice for a very good reason: it is speedy, efficient, and inexpensive. Complaint to award can be concluded in less than 45 days. It is, though, a hermetic procedure in that it applies its own jurisprudence which generally matches U.S. federal law established under the ACPA; but not always and may have something to learn. The tension between the two is expressed in the outcomes of cases challenging UDRP awards. When panelists and judges draw different inferences from the facts and reach different conclusions about liability, it's natural to question which law should be applied; and when the court annuls an award, what the takeaway lessons should be for panelists.
The challenge starts with the jurisdictional scope of the UDRP and the sometimes absent, incomplete, or unprofessional submissions of evidence from respondents that, in some factual circumstances, advantage complainants. Error of reasoning doesn't happen often, but that it happens at all can be disturbing. In its Second Staff Report Paragraph 4.1(c), the Internet Corporation for Assigned Names and Numbers (ICANN) stated that the UDRP was "intended only for the relatively narrow class of cases of abusive registrations." The overwhelming number of cases (in the low 90% range) fall within the "narrow class." These are generally cases of obvious infringement in which facts rather than supposition or conjecture directly implicate respondents' unlawful intentions. However, as facts supporting cybersquatting become less obvious, and there is more dependence on inference, inclusion in the narrow class becomes increasingly attenuated, and sometimes draws in cases that are not in the narrow class.
This is illustrated by a few decisions in which panelists have held respondents liable for cybersquatting that courts have corrected by annulling UDRP awards. Some cases have been decided in trademark owners' favor that should properly have been dismissed as outside the scope of the UDRP. There is a temptation (not always resisted by some panelists, and most certainly unconscious) to give complainants the benefit of the doubt on the issue of bad faith registration. Admittedly, this is a minuscule number compared with the whole, but for losing respondents, there can be a punitive cost for challenging UDRP awards in federal court, even though they prevail. I made an earlier stab at understanding this phenomenon in False Expectations: Attorney's Fees and Statutory Damages in ACPA Actions. If there is any doubt as to the quality of the evidence, a claim should be rejected as outside the "narrow class."
For the most part though, panelists have done an excellent job in identifying and denying complaints outside the "narrow class" — trademark infringement issues, Wisconsin Emergency Medical Technicians Association, Inc. DBA Wisconsin Emergency Medical Services Association, Inc. (WEMSA) v. Marsha Everts, EMS Professionals, Inc., D2018-2841 (WIPO February 7, 2019) — and marks that postdate registration of domain names, Aktsionernoe Obshchestvo, Kontsern Radioelektronnye Tehnologii v. Titan Networks, Domain Hostmaster, D2019-3170 (WIPO February 27, 2020) (<kret.com>). But in other instances where the facts are less complete, where the liability is not obvious, and facts are supplemented with inferences of varying strengths, some panelists give complainants too much credit — Airfx, LLC v. Attn Airfx.com, FA1104001384655 (Forum May 16, 2011) (<airfx.com>, and more recently, discussed below, Lotto Sport Italia S.p.A. v. David Dent, D2016-2532 (WIPO February 13, 2017): both "reversed on appeal" in ACPA actions (I use the terms "reversed" and "appeal" loosely; challenges to UDRP awards in federal courts are not "appeals" but de novo actions). Panelists misconstrue the reach of the UDRP when Procrustes-like they stretch UDRP jurisdiction to include what should properly have been excluded.
There is a sense in rereading some decisions in challenged UDRP awards that they were made ad hoc; in which panelists reached their conclusions without fully parsing or appreciating the facts. The latest example, Lotto Sports is discussed further below. In the earlier Airfx UDRP case, although the domain name predated the trademark, the Panel nevertheless agreed with Complainant that Respondent "did such with the intent to profit by renting such domain names to the legitimate owners of the trademark." As far as one can tell from the Panel's summary of the facts, there was no evidence Respondent could have known about Complainant's future trademark registration, and Complainant offered no evidence of having any common law right that could have made sense of its case.
Admittedly panelists work under great pressure to get their decisions filed (14 days from the date of full submission), but if a case is a square peg, it shouldn't be forced into a round hole. One of the virtues of these "reversals" (and why panelists and parties should pay attention to them) is they teach us to look at facts more carefully and be more skeptical of parties' assertions. The more a decision-maker relies on inference, the greater the likelihood of stepping into error.
If we had to take a count, the majority of challenges to UDRP awards up to now have resulted in settlements in favor of domain name holders. See earlier essay: Prudential Settlements for Alleged Cybersquatting/Reverse Domain Name Hijacking Under the ACPA. Except for a handful of landmark ACPA cases, prominently AIRFX.com v. AirFX LLC, No. CV 11-01064-PHX-FJM, 2012 WL 3638721, at *6 (D. Ariz. June 6. 2013), Black v. Irving Materials, Inc., 17-CV-06734-LHK (N.D. Cal 2019/2020), and now Dent v. Lotto Sport Italia SpA, CV-17-00651-PHX-DMF (D. Arizona Mar. 10, 2020) it has been rare for actions to go the distance to trial and judgment.
What is emphasized in these cases, and the reasoning that goes into the different conclusions, Panel and Court, seemingly from the same facts, is that the court record has been more thoroughly digested (more so than a Panel can devote). It highlights in a dramatic way that in the competition for domain names, trademark rights are inversely impacted by complainants' choices of language; the more common their terms or the use of those terms (even if not common) by other commercial players, the less protection mark owners can expect to enjoy. That a domain name may be identical to a trademark is not evidence of unlawful registration, although it could be with the right ingredients of facts.
The wrong ingredients (the most dramatic situation!) are present in claims by trademark owners who have standing to maintain a UDRP but no actionable claims because their marks postdate the registrations of the domain names. Such registrations cannot, as a matter of law, be unlawful. A recent example is <beautifulpeople.com>, Panel and Court agree on this issue: Beautiful People Magazine, Inc. v. Domain Manager / PeopleNetwork ApS / Kofod Nicolai / People Network Aps / Nicolai Kofod / People Network, FA1502001606976 (Forum May 4, 2015), and Joshua Domond and Harold Hunter, Jr v. PeopleNetwork APS d/b/a Beautifulpeople.Com, Beautiful People, LLC, Greg Hodge, and Genevieve Maylam, No. 17-15576 (11th Cir. Sep. 20, 2018) (attorney's fees against defendant affirmed). The opposite side of this coin is highlighted in another recent case in which the Court expressed skepticism about a domain name plaintiff's claim of reverse domain name hijacking, Pace v. Lundh, No. C18-5965RBL, at *1 (W.D. Wash. Mar. 4, 2020), which I'll return to further below.
In recent months several other losing domain name holders have launched ACPA challenges: Ganeden Biotech, Inc. and Kerry Luxembourg S.à.r.l v. Rob Monster, D2019-3012 (WIPO February 13, 2020) (<BC30.com>); Lakes Gas Co. v. Domain Administrator, DomainMarlet.com, D2019-0830 (WIPO June 21, 2019) (<lakes gas.com>) earlier in 2019 >; and possibly others. In these, however, the sparseness of the records underlines an important point in both UDRP and ACPA actions: inferences from skimpy records are generally drawn against the party having the burden of proof or production. In Ganeden, the Respondent limited itself to an email that stated that "[t]he Disputed Domain Name is generic, and the Complaint should be denied. The Panel should find Reverse Domain Name Hijacking (RDNH)." This is not, I think, a strategically wise submission! In the federal action plaintiff- domain name holder should possibly rethink his submission strategy.
If there is "error" in a Panel's reasoning, it must be in the ultimate finding of bad faith registration. In the Black case, for instance, discussed in more detail in False Expectations previously noted, the domain name holder failed to respond to the UDRP, thus leaving the Pane with a one-sided record. With a full record in the ACPA action, first, an advisory jury held the plaintiff's registration was lawful, then in a lengthy written decision, the Court explained the legal basis for vacating the UDRP award. Unfortunately, the Court denied plaintiff's (losing UDRP domain name holder) motion for attorney's fees, reinforcing the point about false expectations in reverse domain name hijacking cases. Plaintiff vindicated his rights but at a heavy financial cost (absorbed most likely by his attorney in unpaid time charges).
Two new ACPA court decisions were filed earlier in the month. Both should give panelists and parties something to think about. The Pace case noted earlier was probably received with dismay by the challenging domain name holder, while the result in the other was received with joy, Dent v. Lotto Sports, supra. Of the two, Dent is the more important, but Pace illustrates a point that plays a significant role in the outcome of UDRP cases, namely weight of such evidence as has been submitted and credibility. Defendant (trademark winner in the UDRP) did not appear for the ACPA and Pace moved for default judgment, but the court was having nothing of it: "The context of the case suggests serious questions as to the veracity of [Pace's] legal conclusions, under both the ICANN UDRP and the ACPA." I won't go into the case further at this point; rather, will reserve comment until the next motion and ruling.
In Dent, the court had a full record and briefing both of the facts and the law. It concluded that plaintiff's registrations of <lottostore.com> and <lottoworks.com> were lawful. It arrived at this determination by focusing more carefully on two factors that deserve greater attention in UDRP cases since panelists generally view them through a narrower lens, namely 1) domain name creation date; and 2) website content. Ordinarily, panelists commence their analysis of respondent's intention to cybersquat as of the domain name registration date. This works well with cases in the "narrow class," but less well for cases in which the factual circumstances create uncertainty of intention. Remember, two of the bad faith circumstances include the phrase "primarily for the purpose." It can almost be said as a law, as the uncertainty of intention increases the greater the need to rely on inferences, and this may (without conscious intention) lead to giving complainants the benefit of the doubt.
The Court's decision in Airfx demonstrated this with particular clarity. The UDRP Panel viewed the facts through Complainant's eyes; the Court viewed the facts as an outsider piecing together the circumstances to grant judgment to the domain name holder. The Dent facts are different in that the domain names had been acquired from earlier registrants. The creation date predated the trademark, but plaintiff's registration date postdated the mark. This brought into the fact pattern the good or bad faith of earlier holders, which (as I noted) is not ordinarily considered in a UDRP analysis. Domain names predating corresponding but later acquired trademarks are perfectly lawful. The question is whether successors registering domain names after marks are used in commerce have the traditional rights of property ownership? Domain names have been declared by a number of jurisdictions, including the U.S., to be found to be intangible property.
This issue arose in GOPETS Ltd. v. Hise, Digital Overture, Inc., 657 F.3d 1024 (9th Cir. 2011), cited in Dent. Ordinarily, creation date if not important in UDRP adjudication, but the Dent court showed how it can be, and why in this particular case the Panel missed its significance:
Given that the Court has found that Plaintiff purchased the 1998 and 2011 non-party registrations of the disputed domain names, given that there is no evidence of unlawful use or bad faith regarding such registrations, and given that the Court has found that Plaintiff did not use the disputed domain names, the Court need not address bad faith and safe harbor provisions of the ACPA.
The Ninth Circuit had concluded that "in light of traditional property law," there was "no basis in ACPA to conclude that a right that belongs to an initial registrant of a currently registered domain name is lost when that name is transferred to another owner."
Translated to the UDRP, this means the claim was outside the "narrow class," and the complaint should have been dismissed. Whether Dent will suffer the same result on its motion for attorney's fees as in Black remains to be seen, though the facts are slightly different, which makes the prognosis more comforting in Dent.
The second issue in Dent of interest to parties and panelists concerns the use or non-use, or resolution of the domain name to a registrar created, as opposed to a respondent page. Ordinarily, in UDRP proceedings, populated pages are read against respondents. Respondents should take pains to properly curate their website by avoiding infringing links. However, in Dent
Defendant [prevailing party in the UDRP] cites to no case authority discussing or holding that a domain name owner's utilization of a GoDaddy or a similar noncash parking page constitutes "use" of the domain name in the context of a claim under the ACPA. Further, the evidence indicates that: (1) Plaintiff has not developed a public website using the domains; (2) Plaintiff has not advertised or sold any goods or services using the domains; and (3) under his agreements with GoDaddy, Plaintiff has no authority to modify the content on the parked pages and may only inquire of customer support what further options "might be available.
This suggests panelists should dig deeper before holding respondents liable for content created under "noncash parking page" agreements. This is not to say that direct links to complainant or competitors should not be a factor, but the intention to violate the UDRP must be established positively, not by conjecture or for that reason alone.
Whether panelists should look to court decisions for guidance on future UDRP cases is an interesting question, but the Airfx, Black, and Dent judgments "reversing" UDRP awards demonstrate through their analysis and reasoning what the finder of fact should be looking for. In these cases, the UDRP Panels should have dug deeper. These correctives are a positive development. In the same way that appellate rulings establish the law for district court judges, so judges in their reasoning on cybersquatting claims offer a reading of the law for panelists that deserves their attention. The more clearly the boundary of the "narrow class" is fixed, the stronger the UDRP.
Written by Gerald M. Levine, Intellectual Property, Arbitrator/Mediator at Levine Samuel LLP