The WIPO Final Report published in April 1999, from which sprung the UDRP the following October, is useful in shedding light on what the assembled constituencies had in mind in agreeing to particularly contentious issues. One of those issues was whether registrants had to actively search trademark records before purchasing domain names. Other than paragraph 2 of the Policy which codifies registrants’ representations, there is no guidance as to what registrants must do, at a minimum, to verify that their registrations “will not infringe upon or otherwise violate the rights of any third party”, and are not being registered “for an unlawful purpose.”

The phrasing of paragraph 104 of the WIPO Final Report suggests tensions among the constituencies on this issue, and paragraph 105 reflects the compromise:

[104] many commentators stressed the importance of encouraging voluntary domain name and trademark searches, on the part of prospective domain name applicants, to verify that the domain name that they intend to register was unencumbered and did not infringe upon the intellectual property rights of any third-party. It was noted that a range of commercial and public search services existed for both domain names and trademarks. These commentators urged the inclusion, in the domain name application, of language encouraging voluntary searches.

The differences of view were settled with the recommendation in paragraph 105 that “domain name registrations [not] be made conditional upon a prior search of potentially conflicting trademarks, but it is recommended that the domain name application contain appropriate language encouraging the applicant to undertake voluntarily such a search.” Which is all very well, but if registrants merely represent that they will “not knowingly use the domain name in violation of any applicable laws or regulations” what is the representation worth?

Panels have dealt with this conundrum in essentially two ways. Speaking generally, small holders purchasing domain names for personal or business purposes and using them consistent with their semantic meanings prevail. We see this in Hallmark Licensing, LLC v. EWebMall, Inc., D2015-2202 (WIPO February 12, 2016) (<hallmarkgift.com>) and SummerTech Inc. v. Jimmy Hill / iD Tech Camps, FA1512001652542 (Forum February 2, 2016) (<summertechcamp.com> and <summertech.net>). Mega holders are treated differently by being challenged to prove that they have systems in place to avoid infringement. We see this most recently in Blue Ridge Fiberboard, Inc. v. Domain Asset Holdings, LLC, FA160200-1661150 (Forum March 20, 2016) (<soundstop>) which I want to talk about further in a moment.

There is an open question as to whether high-volume registrants should be treated differently and charged with knowledge or awareness of an existing trademark unless proven otherwise. The suggestion that they should is traceable to a 2006 case, Mobile Communication Service Inc. v. WebReg, RN., D2005-1304 (WIPO February 24, 2006) followed shortly thereafter by Media General Communications, Inc. v. Rarenames, WebReg, D2006-0964 (WIPO September 23, 2006) and mVisible Technologies Inc v. Navigation Catalyst Systems Inc., D2007-1141 (WIPO November 30, 2007).

In mVisible Technologies the Panel observed that “a sophisticated domainer who regularly registers domain names for use as PPC landing pages cannot be willfully blind to whether a particular domain name may violate trademark rights,” adding that “a failure to conduct adequate searching may give rise to an inference of knowledge [of complainant’s mark].” The sole Panel in Mobile Communications was the Presiding Panelist in mVisible Technologies.

It can be argued, particularly as trademarks descend to the lower rung of protectability, that imposing a higher duty on high-volume registrants for that category of mark is discriminatory, but it can also be argued that there is good reason to require more than a pro forma representation. Unlike small holders who register domain names for their own business or personal purposes high-volume registrants indiscriminately scooping up expiring domain names cannot possibly in good faith make the representations required by Paragraph 2. This was precisely the point made by the Panel in Mobile Communications:

Where . . . a respondent registers large swaths of domain names for resale, often through automated programs that snap up domain names as they become available, with no attention whatsoever to whether they may be identical to trademarks, such practices may well support a finding that respondent is engaged in a pattern of conduct that deprives trademark owners of the ability to register domain names reflecting their marks.

At the time Respondent in Blue Ridge Fiberboard acquired the domain name it was purchasing names at the rate of 500 a day. At that rate, what has the registrant done to comply with paragraph 2’s injunction that “[i]t is your responsibility to determine whether your domain name registration infringes or violates someone else’s rights”? In this particular case, Complainant has years of priority for SOUND STOP. Is the phrase “sound stop” generic because each word is found in the dictionary (as Respondent argues)? Or, is the phrase suggestive or arbitrary as it applies to the goods Complainant offers? The Panel answered the question by awarding to Complainant. Within the stipulated 10 day stay period of the award, the Respondent commenced an ACPA in the Western District of Washington at Seattle.

Respondent’s practice in selecting domain names to purchase, which it explains in its ACPA action against Blue Ridge Fiberboard, is that it

uses proprietary software and algorithms that analyze available or soon-to-be available domain names, including expiring domain names being sold at auction. Factors the algorithms consider to determine whether a domain name would be of value to Plaintiff’s customers include the number of characters in the domain name, whether the domain name includes common terms, and whether the domain name has features that would make it marketable and easy to remember, such as the use of alliteration.

Notably, there is no algorithm to check the USPTO database for corresponding trademarks. Is this the kind of adequate search called for by mVisible Technologies or does it fall short, and falling short is evidence of bad faith registration and bad faith use?

What’s at stake here is whether the mere existence of a trademark registration should put the registrant on notice (the equivalent, perhaps, of constructive knowledge) such as to support a claim of bad faith. The answer to this question is particularly important where an expiring domain name is scooped up and a trademark owner is prevented from obtaining the corresponding domain name and is holding it for ransom as has happened in another recent case.