Push to Adopt “Renewed in Bad Faith” Standard in UDRP puts Investment Domains at Risk
Many domains held for investment are at risk of loss due to a radical new approach to the UDRP championed by some UDRP panelists at WIPO’s Advanced Workshop on UDRP Practice and Precedent held in Geneva last week.
Respected WIPO Panelist and the Workshop co-leader, David Bernstein, is a leading advocate for this new approach that changes the criteria that the Complainant must prove to win a transfer of a disputed domain. The standard since the UDRP was introduced in 1999 has been that “Registration in Bad Faith” must be proven before a domain can be transferred. Bernstein, and other panelists who share his views, are using a “Renewed in Bad Faith” standard instead.
The “Renewed in Bad Faith” standard is an open invitation to covetous companies to employ the UDRP to try to steal your domains. As was made clear at the WIPO workshop, under the “renewed in bad faith” standard every renewal provides a panel the opportunity to look at the then current use for evidence of bad faith. Putting a domain to virtually any use other than running an established business on that domain can be viewed as bad faith. The following were all held out as examples of bad faith use at the workshop, and have been cited as bad faith use in many UDRP decisions-
● parking a domain name, whether there are infringing links or not;
● having a GoDaddy landing page on your domain, even if you make no money from it and are not aware of it;
● leaving the domain undeveloped;
● offering the domain for sale;
● having a criticism site on a domain;
● running a business on the domain, if the business is viewed as competitive with the Complainant’s.
Renewing your domain while it is being put to any of the above uses therefore could give a panel grounds to order a transfer under the “Renewed in Bad Faith” standard.
Bernstein used the “Renewed in Bad Faith” standard to order the transfer of Big5.com in a controversial recent decision. The decision is so controversial because Big5.com was clearly registered in good faith. The Taiwanese owner originally registered big5.com for a business in connection with the ‘big 5’ Asian languages, and had a trademark on the term ‘Big 5’ for that purpose, but then allegedly used the domain to target the trademark rights of Big 5 Sporting Goods.
The UDRP details the rules governing a domain dispute in Paragraph 4 of the policy where it states that in order for a transfer to be ordered the trademark holder bringing the dispute, the “Complainant”, must prove three elements, including as specified in section 4(a)(iii), that the “domain name has been registered and is being used in bad faith”. Since all parties acknowledged that Big5.com had been registered in good faith, it would seem impossible for the Complainant to have proven bad faith registration.
How can Bernstein, and the other panelists who take this approach, justify disregarding the clear language of the UDRP? They look outside of Paragraph 4 to the previously little noticed Paragraph 2. Paragraph 2 is a warranty between the domain registrant and the domain registrar. In Paragraph 2, the domain registrant warrants to the registrar that “By applying to register a domain name, or by asking us to maintain or renew a domain name registration, you hereby represent and warrant to us that …(b) to your knowledge, the registration of the domain name will not infringe upon or otherwise violate the rights of any third party.”
Those who look to Paragraph 2 see it as placing on the domain owner “a continuing duty to ensure that the domain name is not used in violation of another’s rights”. In their view, “The benefit of an original good faith registration should not be perpetual to the point where it can cloak successors in title and successors in “possession” long after the original registration would have expired.”
In other words, according to those panelists who are relying on the Paragraph 2 language, each time a domain owner renews a domain, he is warranting anew that the domain will be used in good faith. Since most domains are auto-renewed each year, this interpretation has the effect of rendering irrelevant the original registration date of the domain, regardless of whether the domain was registered a decade or more earlier.
The dual requirement under the UDRP that a domain must be proven to be BOTH REGISTERED in bad faith and USED in bad faith as a precondition to transfer, is replaced under this approach by condensing the Registration/Renewal date and the use date so that in most cases they are within one year of each other. For practical purposes the Paragraph 2 standard is simply a USED in bad faith standard, as the registration requirement is thrown out the window.
This is made clear in the majority decision in the Big5.com dispute: “the Panel 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).”
Bernstein and his fellow panelist in the majority decision focused on the recent use of the domain. They found that “the disputed domain name at the time of its last renewal was already a prototypical cyber squatting [site] unrelated to any of Respondent’s original business”. By pulling into the arbitration criteria the renewal warranty from Paragraph 2, they thought they had sufficient justification to order the transfer of the domain.
How one looks at the Paragraph 2 approach is likely a matter of perspective. Seen from the perspective of a trademark owner whose mark is being blatantly infringed by a domain name that was originally registered in good faith but then repurposed to abusive target your mark, the flexibility of the Paragraph 2 approach is attractive.
This new approach results from frustration among many panelists that the UDRP as traditionally interpreted is powerless to combat cybersquatting that happens when domains that are legitimately registered are then used to infringe on trademark rights that arise after the domain was registered.
As I understand it, Bernstein and others who are advocating this approach see it as appropriate when there has been a change in the original use of the domain where the domain owner intentionally and opportunistically uses his domain to target trademark rights that have arisen since the original domain registration. But this nuance is likely to be lost on other panelists who are either not as experienced, or who are more aggressive in their approach. If the Paragraph 2 approach takes hold, then I would expect to see a tidal wave of cases all making variations on the following argument-
“It doesn’t matter that XYZ domain was registered 10 years before my company came into existence, when it was last renewed it was being used as a [parked page/ blank page/ for sale page/ under construction page/ GoDaddy lander page/ criticism site] which is a bad faith use that violates the warranty under Paragraph 2 of the UDRP, therefore transfer the domain to me.”
Bernstein and his colleagues may feel a sense of pride that they have found a clever way to strengthen the UDRP to better fight cybersquatting. If this approach takes hold, however, the true consequence is that they will have gutted the UDRP, putting at risk most domains held by domain investors, and even domains held by small businesses that are no longer actively being used.
As with other attempts to introduce novel interpretations into the UDRP, a well-intentioned effort to more effectively combat abusive cybersquatting can create serious side effects that cause harm that far outweighs any good that is achieved. We saw this most recently with the Octagen/Mummygold series of cases where panelists Andrew Christie and Scott Donahey led an effort to revise the paragraph 4(a)(iii) language cited above: “domain name has been registered and is being used in bad faith” by replacing the “and” with an “or” so that merely bad faith use, not bad faith registration, would need to be proven to satisfy this criterion for transfer.
The Octagen/Mummygold line of reasoning encouraged a slew of frivolous UDRP Complaints from companies that wanted to seize domains registered long before the Complainant companies came into existence. As described in the “John Berryhill’s UDRP University” post, Berryhill’s frustration with this line of reasoning led him to select Christie for the panel of these frivolous cases so that Christie could see first hand the consequences of his tampering with the UDRP.
According to Bernstein and Tony Willoughby, his Workshop co-leader, the consensus view of the community of UDRP panelists is now to reject the Octagen/Mummygold approach, in part because to adopt it would undermine 14 years of UDRP precedent. For panelists wishing to maintain the consensus view, Octagen/Mummygold is now foreclosed as an approach for dealing with “repurposed” cybersquatting, leaving panelists in search of a new approach.
What is attractive about the Paragraph 2 approach to its advocates is that it is a novel interpretation that, in their view, does not undermine established precedent as it hasn’t been fully considered before. Yet the Paragraph 2 approach strikes me as disingenuous, because it is simply the Octagen/Mummygold approach in new clothing. While the adoption of Paragraph 2 warranty language relies on a different provision and a slightly different line of reasoning, the practical effect is nearly identical to the Octagen/Mummygold approach, as both approaches permit a panel to disregard the requirement that bad faith registration must be proven. If this approach is illegitimate under Octagen/Mummygold because it undermines 14 years of precedent, it is similarly illegitimate under the Paragraph 2 approach. The Paragraph 2 approach is a new bite at the same apple, a similar violation of UDRP core principles that would also undermine 14 years of precedent. It is simply another way to change the “and” to an “or”.
I find it very concerning that David Bernstein is a leading advocate of the Paragraph 2 approach, as he is held up as an expert on the UDRP and entrusted by WIPO with instructing other UDRP panelists on how to implement the UDRP, so he is very influential. Also, unlike some other panelists, Bernstein is not considered to be hostile to domainers. On the contrary, he has been seen as a panelist who is willing to give a fair hearing to domain investors. Ari Goldberger often selected Bernstein for the disputes he handled. Earlier this year, I advised a domain owner who was handling his own response to consider selecting Bernstein to be on his 3-member panel. Bernstein has shown a willingness to slap Complainants with a finding of Reverse Domain Name Hijacking. He is associated with more RDNH decisions than all but two or three other panelists.
The paragraph 2 approach has also been adopted by Richard Lyon, another well respected and very experienced UDRP panelist who is considered to offer an even-handed treatment to domain investors. Lyon decided the controversial Sporto.com where he ordered the transfer of a 13-year-old domain that had been registered in good faith because of recent use that he found had been in bad faith.
The move to use Paragraph 2 language cannot be attributed to panelists who are “out to get” domain investors. Rather I think these panelists are legitimately concerned that the UDRP as traditionally interpreted does not adequately address certain instances of blatant cybersquatting and, as described above, the panelists are looking to the paragraph 2 language to cover a perceived “gap” in the UDRP’s ability to address cybersquatting. Unfortunately their proposed “fix” breaks the UDRP.
The Paragraph 2 approach is not universally accepted, and has its critics, even among other WIPO panelists. Bernstein’s co-leader, Tony Willoughby, indicated that he is not persuaded that the Paragraph 2 analysis is correct because it requires a panelist to go outside of the confines of Paragraph 4 where the criteria to be used in a UDRP dispute are specified. One of the pleasures of attending the workshop was the opportunity to spend time with John Berryhill, his wife MJ Keukelaar, Paul Keating and Zak Muscovitch. They all view the Paragraph 2 approach as illegitimate.
Their reasoning as to why Paragraph 2 language is not relevant for UDRP disputes relies on how to properly read and interpret legal documents, an area in which I have no expertise. As best as I understand it, Berryhill, Keating and Muscovitch rely on several arguments. First, the UDRP exists as a function of contract (between the registrant and registrar). The UDRP dispute procedures are incorporated within that contract. Paragraph 2 was taken from the original registration agreement. It consists of a series of representations and warranties by the registrant to the registrar, the only parties to the agreement. This is obvious by the use of the pronouns “you” and “we”. A third-party trademark holder has standing only with respect to the UDRP dispute procedures outlined in Paragraph 4.
Second, the UDRP itself is limited to the terms of the Policy. Paragraph 4 of the Policy specifies the limits of the resolution mechanism and details the three criteria that must be proved for a complainant to prevail in a UDRP. Indeed, as Berryhill points out, paragraph 4 starts with “This Paragraph sets forth the type of disputes for which you are required to submit to a mandatory administrative proceeding. “ In other words, the dispute mechanism is limited to paragraph 4. Paragraph 4 does not incorporate Paragraph 2. Further, Paragraph 4 uses the term bad faith registration. It does not reference “renew or maintain.”
The history of the UDRP drafting process supports a conclusion that paragraph 2, itself a late addition to the document and a sloppy cut-and-paste from the registration agreement cannot be used to drastically expand the mechanism carefully set out in Paragraph 4 so as to render “renew and maintain” equal to “register”.
Berryhill also cites the First WIPO Report on the UDRP that makes clear that “the scope of the administrative procedure [is] limited to the abusive registration of domain names.” Berryhill’s analysis demonstrates that the push to rely on Paragraph 2 warranty language cannot be justified.
Both Keating and Berryhill also point out some of the problems in trying to incorporate a reference to “renew and maintain”. For example, domain names are now renewable for up to 10 years. How does a panel know when the last renewal took place for the purposes of applying this new standard? Even more problematic is the reference to “maintain”. As both point out, a registration is by definition maintained throughout its registration period. As such, incorporating “maintain” effectively renders both registration and renewal meaningless and results in “use” becoming the sole and continuing standard for measuring bad faith.
Keating takes it even one step further. He argues that even if one were to accept that Paragraph 2 creates a standard, the very language is limited to the actual knowledge of the registrant. Indeed the language is “to your knowledge, the registration of the domain name will not infringe upon or otherwise violate the rights of any third party.” The words measure actual knowledge without a duty to investigate. The problem, Keating notes, is that most panelists already tend to ignore this by favoring an obligation to investigate. According to Keating, the entire Paragraph 2 contingent are thus opening the door to a strict liability approach in which all trademarks are elevated to the level of being famous and you and I are obligated to dig them up to avoid the appearance of conflict.
It is still too early to see how this split view on Paragraph 2 will play out. Perhaps the Paragraph 2 approach will become the consensus view, where all panelists will feel an obligation to follow it. If so, domains held for investment will be at high risk of loss and it may be too risky to continue parking investment grade domains. It also would mean that domainers must undertake an investigation every time a domain is renewed so as to ensure no conflict exists with a trademark.
Perhaps like the Octagen/Mummygold approach, the attempt to rely on language from Paragraph 2 will eventually be decisively rejected. But in the meantime, with leading panelists including Bernstein and Lyon advocating for it, we can expect to see many decisions taking the Paragraph 2 approach, and many domains being lost despite having been registered in good faith.
In the succeeding months, as the Paragraph 2 approach becomes better known, we are almost certain to see trademark holders filing complaints in attempt to seize valuable domains that pre-date their existence. A juicy target for these complaints will be domains that are parked on landers that are auto-optimized, as the optimization algorithm can focus the links on the trademarked use. A panel relying on the warranty language from Paragraph 2 could find a domain owner responsible for these advertising links that appear on a parked page even though the links are automatically supplied by Google and the domain owner has no knowledge of them. The panel can determine that the latest renewal while the domain was being parked was in bad faith and order the transfer of the domain. The domain owner will then find that a domain that she registered in good faith many years before the trademark owner came into existence has been lost in a UDRP dispute.
What is even worse is the ability of panelists to find bad faith based on Paragraph 4(b)(iv) (intentionally attracting for commercial gain ) which does not require actual conflicting links. Any commercial use that is not purely descriptive can be considered a bad faith effort to trade on the goodwill belonging to the trademark holder under this standard. This gives panels a rationale to find bad faith for a broad range of uses, and makes it even easier to separate a domain owner from her domain.
A major practical problem with the divergent views on the Paragraph 2 approach is the uncertainty it introduces into administration of the policy. As domain investors, we know all to well the lack of consistency among UDRP panelists in how they approach bad faith use. Some panels have found that earning revenue from PPC links is a legitimate use for a domain, others have found that evidence of bad faith. Some panels have found that the business of selling descriptive domains – whether or not there are trademarks on the descriptive word on which they are based – is a legitimate use for a domain, some panels have found that to be bad faith. Some panels have found that not developing a domain after owning it for many years no cause for concern, others find that evidence of bad faith. As there is no consistent standard for what is consider use in bad faith, domain investors rely on the requirement that before a domain can be ordered to be transferred that registration in bad faith must be proven. This is the pillar on which we rest our defense in a UDRP response in the absence of a consistent standard for bad faith use. The paragraph 2 approach knocks this pillar out from under us. When registration in good faith no longer is a sufficient defense, then the UDRP truly becomes a crapshoot. The decisive issue of what constitutes bad faith varies from panelist to panelist.
While there is disagreement about whether the incorporation of “renew and maintain” language from Paragraph 2 belongs in the UDRP dispute mechanism laid out in Paragraph 4, there can be no disagreement that introducing this radical approach into the UDRP destroys any consistency of application of the UDRP. The disregard that Bernstein, Lyon and others show for the principle and the value of consistency in application of the UDRP is shocking to me. They must be aware that adopting this approach is like throwing a bomb into the shaky edifice of UDRP precedent and the 14 year effort to develop a consensus approach to UDRP interpretation, as this approach blows that whole effort into smithereens.
What I have seen is that faced with perceived shortcomings in the Policy, it becomes very attractive to certain panelists to experiment with ways of expanding the Policy beyond the limited scope that was agreed to after extensive negotiations during the creation of the Policy. These panelists are no longer content with implementing the policy. They want to make policy. Because they are entrusted and empowered to implement the policy according to their own interpretation, without any oversight, they are free to creatively “interpret” the Policy in ways that result in profound changes to the Policy.
It is likely not a coincidence that many of these activist panelists are concurrently trademark lawyers. They have spent their careers aggressively advocating on behalf of trademark holders. When aggressive advocates are then placed in the role of neutrals, it is no surprise that they don’t exercise the restraint and the judicial temperament that is required for the role of panelist. They keep their advocate hats on.
These panelists will claim that they are honoring the spirit of the Policy and furthering its aim of combating cybersquatting. Instead, they are usurping the policy making function that should develop from deliberations within ICANN and among its many stakeholders.
As we in the domain investment industry are shut out from the process and from the power to make change, we are left with the art of persuasion. Having seen Berryhill, Keukelaar, Keating, and Muscovitch in action last week, they are very actively engaged in this uphill struggle.
I am grateful to John Berryhill, MJ Keukelaar, Paul Keating, and Zak Muscovitch for generously sharing their expertise and feedback, and for fighting for the rights of domain registrants even when we aren’t aware that they are watching our backs.
Revised November 8th