ICA has just submitted its sixth comment letter to ICANN in regard to the ever-evolving Applicant Guidebook (AG) for the new gTLD program.
The latest version of the AG contains a new provision, tucked away in the evaluation questions for new gTLD applicants, that asks whether they will adopt policies to eliminate or minimize “social costs” and “negative consequences”, and then inquires: “Will you impose any constraints on parked sites, or sites that offer only advertising?”
While we have no idea where this question came from or why it has just shown up in what may (hopefully) be the final draft of the AG, we were not going to let it pass unchallenged. Our comment letter demands that ICANN delete this totally unjustified criteria before the Guidebook is finalized. ICA’s letter points out that ICANN, by its own admission, is not in the business of dictating business models or content, and that domain monetization through “parking” in and of itself is a completely legal activity that can provide substantial consumer benefits. What’s more, ICANN’s own policy paper on domain monetization, the proposed rules for determining bad faith under Uniform Rapid Suspension (URS), and WIPO’s latest guidance for UDRP examiners all recognize that non-infringing parking is a permissible activity well within the legitimate rights of a registrant to use his domain for any lawful purpose.
ICANN might be better concerned with whether any new gTLD will attract enough type-in traffic to make domain parking a viable option – indeed, with whether most new gTLDs will attract sufficient registrants to succeed. The outcry last week over suggestions that URS be applied to .Net suggests that knowledgeable registrants may well avoid all new gTLDs where domains can be suspended through a $300 proceeding. Not surprisingly, comments filed by trademark interests on the latest AG propose a further tightening of URS rules – including lowering the number of domains threshold for triggering “loser pays”, diluting the evidentiary standard, eliminating the requirement to establish bad faith, adding a domain transfer option to make URS the functional equivalent of the UDRP, and broadening the range of trademarks and phrases containing trademarks vulnerable to a URS action. Acceptance of those modifications could well result in most new gTLDs being .Fail and .DOA.
ICA’s letter does commend ICANN for accepting some of the suggestions made in our December 9, 2010 comment letter on the last version of the AG, while continuing to press them to make reasonable modifications in other areas. We also urge that the Board cast a vote to launch the new gTLD program when it meets in Singapore next month – if for no other reason than to halt the additional erosion of registrant rights that is likely to result from any further extension of this AG development process.
The full text of ICA’s letter follows —
VIRTUALAW LLC Philip S. Corwin, Founding Principal 1155 F Street, NW Suite 1050 Washington, DC 20004 202-559-8597/Direct 202-559-8750/Fax 202-255-6172/Cell email@example.com
May 15, 2011
By E-Mail to
Board of Directors
Internet Corporation for Assigned Names and Numbers (ICANN)
4676 Admiralty Way, Suite 330 Marina del Rey, CA 90292-6601
Re: New Generic Top-Level Domains (gTLDs) Applicant Guidebook – April 2011 Discussion Draft
Dear Members of the ICANN Board:
This comment letter is submitted by the Internet Commerce Association (ICA) in regard to ICANN’s April 15th notice (http://icann.org/en/topics/new-gtlds/comments-6-en.htm) establishing a period for public comments on the April 2011 Discussion Draft of the Applicant Guidebook (AG) for new gTLDs. This letter supplements all of our comments on prior versions of the proposed Guidebook, the most recent of which was filed on December 9, 2010 (http://forum.icann.org/lists/5gtld-guide/msg00052.html).
ICA is a not-for-profit trade association representing the domain name industry, including domain registrants, domain marketplaces, and direct search providers. Its membership is composed of domain name registrants who invest in domain names (DNs) and develop the associated websites, as well as the companies that serve them. Professional domain name registrants are a major source of the fees that support registrars, registries, and ICANN itself.
• New AG language that equates parked domains with negative social consequences and costs is unjustified, inappropriate, and at odds with prior ICANN findings and policy, proposed URS evaluation criteria, and WIPO guidance to UDRP examiners. We demand its deletion.
• The inflexible “three strikes” UDRP reference for new gTLD applicant disqualification remains unjustifiably inflexible and should be modified. If retained, we shall be monitoring whether it is equivalently applied to – or equivalent flexibility is accorded – corporate new gTLD applicants, such as major registrars, and individual members of investment groups.
• We appreciate ICANN’s responsiveness to our suggestions that UDRP decisions reversed upon appeal not count against applicants, and that serial reverse domain name hijackers receive the same treatment as multiple UDRP losers.
• We are disappointed that ICANN has retained the inadequate 14-day response time to URS complaints, and has failed to provide greater clarity regarding criteria for approving good faith extension requests. We urge reconsideration.
• We are gratified that the revised version of the URS does not contain a domain transfer option and urge adherence to that position. We also urge that the ICANN Board reject any further GAC requests to further disadvantage registrants through additional URS modifications – and urge that the AG development process be brought to a conclusion with a vote for the launching of new gTLDs at the upcoming Singapore meeting.
Unjustified Bias Against Domain Parking
While the majority of our comments in this letter will update the positions expressed in our December 2010 comment letter to reflect alterations and additions in the current discussion draft, we were dismayed to see that the latest version of evaluation criteria for new gTLD applicants (http://icann.org/en/topics/new-gtlds/draft-evaluation-criteria-redline-15apr11-en.pdf) indicates an unexplained and unjustifiable bias against domain “parking”. This evaluation criteria is completely at odds with prior ICANN descriptions of domain parking, as well as with its treatment under the proposed Uniform Rapid Suspension (URS) mechanism and with current WIPO guidance for UDRP examiners. This inappropriate intrusion into a registrant’s right to utilize his domains for all lawful purposes is also at odds with current marketplace realities.
Question 18(c) in the Attachment to Module 2 states the following –
What operating rules will you adopt to eliminate or minimize social costs (e.g., time or financial resource costs, as well as various types of consumer vulnerabilities)? What other steps will you take to minimize negative consequences/costs imposed upon consumers? Answers should address the following points:
…iv. Will you impose any constraints on parked sites, or sites that offer only advertising? (Emphasis added)
This question clearly would impose pressure on applicants to impose rules that would limit parked websites at domains registered within their new gTLD, and equates domain pa
rking with “social costs” and “negative consequences/costs”. Such derogatory association is unfounded, and registry-imposed constraints are unjustified and unnecessary.
Our starting point is our belief that a domain registrant, having paid the annual registration fee for a domain, is entitled to engage in any lawful activity with that domain – or to not use it at all. There is no scarcity of domains available for use under the present gTLD and ccTLD regime, and any allegations of scarcity will lose all credibility once dozens or hundreds of new gTLDs commence operation. Criticism of domain parking is generally misinformed, incorrectly equates parked pages with trademark infringement, and generally boils down to jealousy that a registrant has obtained valuable generic domains at a highly trafficked TLD before the critic realized the inherent value of such domains.
ICANN’s 2008 Policy Issue Brief on Domain Name Monetization (http://www.icann.org/en/policy/briefs/domain-name-monetization-jun08-en.pdf) both described domain parking and recognized that it is generally non-problematic:
“Domain parking is a practice used by registrars, individual registrants and Internet advertising publishers to monetize type-in traffic. Type-in traffic refers to Internet users who visit a web-page by typing its URL directly into their browser rather than by clicking on a link from another page such as a search engine result page. Proponents of domain parking say it uses domain names to deliver relevant advertising and enhanced search options instead of serving Internet users with an error page often referred to as a 404 file not found.”
With parking, the Internet user is taken to a landing page that has links that generate income for the person or company hosting the site similar to the PPC [Pay-per-click] model described above. Traffic can come from several sources, including out-of-date bookmarks, people typing in the registered domain name, or search engine results. A registrant, for example, may register a domain name but may not be ready to establish a website yet. The registrar then parks the registrant’s domain name on a website that features paid links. Each time Internet users try to locate the website that would correspond to the domain name, they will be directed to the parked page that features an array of advertising links, often offering goods and services related to the domain name. When the user clicks on any of these links, it generates revenue for the registrar, and depending on the terms of the registrar’s parking program, the registrant as well. Some estimate that large registrars earn millions of dollars each year from the PPC fees they earn on parked websites…The practices of using parking and PPC to monetize domain names have not to date generated the same concerns as tasting.
Since that Issue Brief was published ICANN has adopted new policies – supported by ICA – that have essentially eradicated abuse of the add/grace period (AGP) and thereby eliminated abusive domain “tasting” and the “typosquatting” domains that often resulted from its use. It is well know that advertising is the economic basis of much of the commercial activity on the Internet, and domain parking is a perfectly legitimate technique for making consumers aware of products and services associated with a generic word through the direct search monetization model utilized by many domain investors and associated with natural type-in traffic. The display of ad links on non-infringing generic word domains is equal in legitimacy and usefulness to the paid ad links displayed by major search engines when the same word is entered into them.
This new evaluation criteria is also completely at odds with the recognition elsewhere in the April 2011 Discussion Draft that domain parking in and of itself is not a negative criteria under the proposed URS. The latest version of the URS (http://icann.org/en/topics/new-gtlds/draft-urs-redline-15apr11-en.pdf) states–
5.9 Other factors for the Examiner to consider:
5.9.2 Sale of traffic (i.e. connecting domain names to parking pages and earning click-per-view revenue) does not in and of itself constitute bad faith under the URS. Such conduct, however, may be abusive in a given case depending on the circumstances of the dispute. The Examiner will take into account:
188.8.131.52. the nature of the domain name;
184.108.40.206. the nature of the advertising links on any parking page associated with the domain name; and
220.127.116.11. that the use of the domain name is ultimately the Registrant’s responsibility (Emphasis added)
The World Intellectual Property Organization’s “Overview of WIPO Panel Views on Selected UDRP Questions, Second Edition (“WIPO Overview 2.0”)” (http://wipo.int/amc/en/domains/search/overview2.0/), just issued on March 31, 2011, adopts a similar view for the guidance of examiners in the Uniform Dispute Resolution Policy (UDRP) arbitration dispute process. The relevant portion of the Overview states –
2.6 Do parking and landing pages or pay-per-click links generate rights or legitimate interests in the disputed domain name?
Panels have generally recognized that use of a domain name to post parking and landing pages or PPC links may be permissible in some circumstances, but would not of itself confer rights or legitimate interests arising from a “bona fide offering of goods or services” [see also paragraph 3.8 below] or from “legitimate noncommercial or fair use” of the domain name, especially where resulting in a connection to goods or services competitive with those of the rights holder. As an example of such permissible use, where domain names consisting of dictionary or common words or phrases support posted PPC links genuinely related to the generic meaning of the domain name at issue, this may be permissible and indeed consistent with recognized sources of rights or legitimate interests under the UDRP, provided there is no capitalization on trademark value (a result that PPC page operators can achieve by suppressing PPC advertising related to the trademark value of the word or phrase). By contrast, where such links are based on trademark value, UDRP panels have tended to consider such practices generally as unfair use resulting in misleading diversion.
Thus, ICANN’s prior examination of domain parking, the relevant language of the URS, and WIPO guidance for the UDRP all lead to the conclusion that non-infringing domain parking is not problematic for consumers, not indicative of bad faith use, and is permissible and consistent with the recognized rights or legitimate interests of a domain registrant. Hence, there is no justification for any applicant evaluation criteria that equates domain parking with negative social consequences or costs.
In addition, such evaluation criteria would impose duties on new gTLD registries and the registrars authorized to sell and renew their domains that are inappropriate as well as inconsistent with the Internet’s development to date. Presumably, “constraints on parked sites” would require registrars to make inquiries into a registrant’s intended use of a domain, and impose some duty on a registry to review all domains within its registry for compliance with such constraints. Such domain content review is completely inappropriate absent evidence that a registrant is making infringing or illegal use of a website. It would also be trivial to route around by simply adding some minimal amount of content to a domain so that it was not solely populated by advertising links. ICANN has repeatedly emphasized that it has no intention of engaging in content regulation – as was recent
ly evidenced by the Board’s approval of a contract for the .XXX “adult content” registry – but a policy that discourages a certain form of perfectly legal commercial speech at new gTLDs is a form of content regulation – as well as an attempt to dictate permissible business models — and is therefore thoroughly outside of ICANN’s jurisdiction and unacceptable.
Further, the present use of domains indicates that parked pages are quite common yet there is no evidence that they impose any negative burden on the Internet or its users. VeriSign’s February 2011 Domain Name Industry Brief (http://verisigninc.com/assets/domain-name-report-feb-2011.pdf) reports that 21% of 104.6 million domains that it analyzed are single page websites and further states: One-page websites include under construction, brochure-ware and parked pages in addition to online advertising revenue generating parked pages. Compared to the more than 21 million single page domains identified by that analysis, millions of which are almost certainly parked pages containing multiple ad links, (in a survey of only half of all registered domains) there were less than 6,000 UDRP actions initiated in 2010. So there is clearly no pervasive linkage between domain parking and trademark infringement. In fact, to the extent that non-infringing parked pages are profitable, that constitutes proof that they are providing benefits to consumers who find value in their links.
Finally, in addition to be being unjustified and inappropriate, registry-imposed constraints on domain parking at new gTLDs are unnecessary given marketplace realities. Domain industry investors are generally skeptical about the prospects of success for new gTLDs and are unlikely to aggregate significant portfolios of parked domains in them. Purely parked pages are unlikely to rank high in search engine results, and such domains at new gTLDs for the large part will attract little in the way of the natural type-in traffic required for profitable parked page monetization. Finally, there has been a shift in the last few years within the domain investment industry from parking to development as PPC rates decline and investors become more sophisticated in their utilization of valuable generic domains. In sum, even without the adoption of a highly objectionable anti-parking policy, parked websites are likely to constitute a significantly smaller proportion of registered domains at new gTLDs than they have at incumbents.
For all of the above reasons, we demand that ICANN remove clause iv from Question 18(c) in the Attachment to Module 2.
UDRP “Three Strikes” Rule for New gTLD Applicants
Our December 2010 comment letter urged that “The inflexible “three strikes” UDRP loss applicant disqualification criteria must be adjusted to take into account the size of an applicant’s domain portfolio as well as their overall history of UDRP arbitrations.” Notwithstanding that request, Section 1.2.1.m of the April 2011 AG still establishes a policy of automatic disqualification, absent “exceptional circumstances”, for any entity with or including any individual that –
has been involved in of a pattern of adverse, final decisions indicating that the applicant or individual named in the application was engaged in cybersquatting as defined in the UDRP, ACPA, or other equivalent legislation, or was engaged in reverse domain name hijacking under the UDRP or bad faith or reckless disregard under the ACPA or other equivalent legislation. Three or more such decisions with one occurring in the last four years will generally be considered to constitute a pattern. (New language underlined)
We continue to believe that the “three strikes” criteria is too inflexible and that applicant evaluation criteria should take into account the total size of an applicant’s domain portfolio as well as the percentage of adverse UDRP decisions rendered against them in comparison to all UDRP proceedings they have been involved with.
Once the new gTLD application window opens we shall be carefully monitoring the actual enforcement of this disqualification criteria for both companies and individuals, if it remains unchanged. If strictly applied to companies – by being made applicable to new affiliate or subordinate entities they have established and control for the sole purpose of constituting the applicant for a new gTLD – it will bar new gTLD applications by many of the largest domain registrars. This result would seem highly improbable, especially in light of the AG’s relaxation of registry-registrar separation requirements. Hence, we suspect that ICANN will exercise a significant degree of enforcement flexibility based upon the “exceptional circumstances” and “generally be considered” phrases in the current proposed language, rather than bar applications for new gTLDs submitted by large registrars.
We do not advocate the result of barring applications by such registrars or other corporate entities – we simply wish to make the point that any flexibility in applying these criteria should be accorded equally to both corporate and individual new gTLD applicants. Individual owners of large domain portfolios, unlike corporations, cannot establish subsidiaries or affiliates of themselves, yet may otherwise be in the same position as regards their UDRP histories.
Recognition of Court Reversals of UDRP Losses
Our December 2010 comment letter took the position that “A UDRP loss that has been reversed upon appeal to a court of proper jurisdiction should not be counted against an applicant.” The new language in Section 1.2.1.m referencing “adverse, final” decisions appears to be responsive to that suggestion and we thank ICANN for that adjustment.
Equivalent Treatment of Reverse Domain Name Hijackers
Our December 2010 comment letter urged, as a matter of equity, that “Applicants that have been found to have engaged in three or more instances of attempted reverse domain name hijacking should be barred from participation in the new gTLD program.” New language in Section 1.2.1.m that disqualifies entities who have engaged in RDNH under the UDRP, or bad faith or reckless disregard under the ACPA or equivalent legislation, is responsive to our suggestion and we thank ICANN for that adjustment.
URS Response Time and Good Faith Extensions
Our December 2010 comment letter advised that “The STI-RT’s recommendation for a 20-day URS response time should be restored or, failing that, extensive guidance should be provided regarding the grounds for which a 7-day “good faith” request for response extension will be granted.” We are disappointed that Sections 5.1 and 5.3 of the current URS draft are not responsive to this request and again urge that, at a minimum, additional guidance be provided in regard to acceptable grounds for a good faith extension.
Additional URS Issue
At the outset, ICANN should recognize that the URS and other new rights protection mechanisms (RPMs) for new gTLDs are likely to be a significant burden on their economic prospects. The Trademark Clearinghouse and Claims Service is likely to result in untold numbers of applicants for generic domains, with no intent to use them for infringing or unlawful purposes, to nonetheless be advised that the domain name they seek may give rise to a URS or UDRP proceeding because the word has been registered by some trademark owner for some commercial purpose somewhere. Most
domain applicants are unsophisticated in their understanding of trademark law and unaware of the limited scope of a trademark, and will likely abandon their registration attempt at the new gTLD rather than proceed and risk being the potential target of an infringement action.
As for more sophisticated applicants, they are likely to avoid new gTLDs to a large extent rather than risk losing the use of their domain in a URS action initiated by a 500-word complaint accompanied by a $300 fee. The massive outpouring of protests just filed in response to the Intellectual Property Constituency’s suggestion that URS be made applicable to .Net as a condition of its registry contract renewal (see http://forum.icann.org/lists/net-agreement-renewal/) is indicative of the disdain in which this untested RPM is held by knowledgeable registrants.
Two significant changes have been made in the URS since the last version of the AG, largely in response to continuing pressure from the Governmental Advisory Committee (GAC) on behalf of trademark owners:
• Section 5.2 now establishes a limited “loser pays” regime. In cases in which a complaint lists 26 or more disputed domains held by a single registrant, the registrant will be required to post a response fee of no more than $300, with that fee going to the prevailing party. While UDRP cases listing more than two dozen names have been rare, and while the capped fee is modest, we must note that even the Implementation Recommendation Team (IRT) opposed establishing a loser pays regime within the URS. If this change is retained, we strongly urge that no reduction in the number of domains giving rise to a loser pays requirement be contemplated. We also must express dismay at the thought that more than two dozen domains could be suspended through an action at a per domain cost of just over $11 – no one can possibly believe that due process can be implemented at that bargain basement price.
• Section 6.4 has cut the time for registrant appeal in default cases from two years to six months, with the possibility of a six-month extension. Even with this reduction any registrant who believes he has been the victim of a wrongly decided URS should have sufficient time to obtain a de novo appeal.
In addition, during the recent discussions between the Board and GAC in Brussels, the Board indicated that it was acquiescing to the GAC request that domain transfers be facilitated via the URS by giving the complainant first option to acquire a suspended domain. On March 3rd we dispatched a letter to Chairman Thrush and CEO Beckstrom that took strong issue with that decision, stating –
However, we are very concerned that the Board has accepted the principle that a rights owner prevailing in a URS has first option to obtain a suspended domain at the conclusion of its registration period. This position is at odds with the IRT’s vision of the URS as a supplement to, and not a substitute for, a UDRP action – that a successful URS action would lead to domain suspension, while domain transfer would remain the exclusive province of the UDRP. Given the far lower cost and faster speed of the URS, and its lessened registrant notice period and rights, we are afraid that many trademark owners will choose to utilize it rather than the UDRP in attempts to obtain domains for their own portfolio. The only significant difference between prevailing in a URS versus a UDRP action is the different evidentiary standard – but, as Chairman Thrush noted on Wednesday morning, it is only “slightly higher” and may well be outweighed for complainants by the URS’ cost and speed advantages. We are also concerned that the availability of domain transfer via the URS will undermine the UDRP reform effort that was recently initiated by the GNSO.
In the alternative, we would suggest that a domain suspended via URS decision simply be placed on a list of domains permanently ineligible for re-registration. Doing so would be entirely consistent with the original aim of having the URS available only for “slam dunk” cases of abuse, as such domains would be unlikely to survive future URS challenges. We believe that such a list would be readily implementable by registrars.
At the subsequent ICANN meeting held in San Francisco, many speakers at the session on RPMs for new gTLDs spoke out strongly against that decision as an open invitation to abusive domain hijackings through the URS.
We are pleased to see that the latest version of the URS does not include such a transfer option and urge that the UDRP remain the sole means by which a complainant can obtain possession of a disputed domain. If this issue should again arise we continue to urge that the concerns of registrants and complainants be accommodated by the win-win approach of placing suspended domains on a permanent ineligible for re-registration list.
Finally, as discussions between the Board and GAC continue on the final elements of the AG and URS, we continue to urge that the Board stand fast against GAC requests that are at complete odds with the community consensus that was reached on RPMs. GAC requests that should be firmly rejected include:
• Replacing the clear and convincing evidence standard with one of preponderance of the evidence as this would again erode critical distinctions between the URS and UDRP.
• Removal of the requirement that a complainant establish registrant bad faith as this would result in a URS having a critical evidentiary element that is lower, not higher, than the UDRP requirements.
• Permanently barring individuals or entities who had lost five URS actions from mounting any defense in future actions against them as no registrant should ever be barred from presenting a valid defense in a singular case.
• Expanding the availability of URS actions to “trademark + keyword” cases as these are not the “slam dunk” cases the URS was purportedly designed to address.
As ICANN has clearly taken many of our past comments into consideration as it made adjustments to this version of the AG, we expect that similarly serious consideration will be given to our input today. In particular, ICANN simply must remove the ill-considered policy against domain parking that has arisen out of nowhere and has no plausible justification.
We appreciate the opportunity to comment on this latest version of the AG – and we certainly hope that it is the final version. Whatever its flaws, it is time to bring the process of establishing the rules governing new gTLDs to a close and commence with the opening of the application period, with additional revisions to new gTLD policy being made in the future based on actual experience rather than overhyped projections. Unfortunately, any continuation of this process is only likely to provide additional time for the GAC to pursue its ill-considered goal of eroding registrant rights on behalf of large corporate trademark interests.
Thank you for your consideration of our views on this matter.
Philip S. Corwin
Counsel, Internet Commerce Association
Comments are closed.