Monday 7 September 2009

The IRT Proposals of an IP Clearinghouse Dangers/Problems And Innovative Solutions

Paper shared on August 7, 2009

Introduction

Under the IRT’s proposal, the IP Clearinghouse takes ICANN far outside its scope and technical mandate. It is an intellectual property structure that has no barrier to creation in the private sector and is consistent with the types of private sector services, including monitoring and alert, now being offered. Moreover, the creation of such a superstructure of ICANN, a single monolith, will attract such lobbying from brand owners that the rollout of new gTLDs may be further delayed.

What the IRT specifically proposes:

· The IP Clearinghouse must be capable of holding data relating to the legal rights of trademark owners, including both registered rights and unregistered rights.

· The data should be submitted by trademark owners directly, or through a registry or registrar, to the IP Clearinghouse together with a reasonable fee.

· Trademark owners must grant a non-exclusive, royalty-free, sublicensable license to their data to ICANN, which will in turn sublicense it to the IP Clearinghouse.

(I think we should move Dangers/Problems up here)

What we recommend: Market-based Private, Regional Trademark Clearinghouses (TMC) - following the WTO model of regional Clearinghouses for trade agreements.

Why opt for Regional Trademark Clearinghouses (TMC):

1. Regional clearinghouses can relate better and respond more efficiently to the differences of trademark laws across the world;

2. Regional clearinghouses will be able to respond and better protect signs of cultural diversity and cultural significance. (Words like ‘the Acropolis’ or ‘the Taj Mahal’ do not have the same historical significance for Greece or India as for the rest of the world. This will also be in line with WIPO’s ongoing work on ‘Traditional Knowledge, Genetic Resources and Traditional Cultural Expression/Folklore”).

3. Regional clearinghouses will work better with national Trademark Offices, which, essentially, are the most appropriate sources to guide this discussion – it is not ICANN. It is the national Trademark Offices that can best work together on a regional basis to support regional TMCs. Further, virtually all national Trademark Offices nowadays maintain databases with their trademark listings. While only some provide access to these databases openly to the public online, they could make access easily available to private TMCs in which they had a hand in creating and/or running and/or overseeing.

4. Regional clearinghouses will be less responsive to abuse, due to their familiarity with the validity of trademark registrations.

5. Regional clearinghouses will also tackle better the concerns over IDNs. To serve IDNs, the private, regional TMCs are in the best position to meeting the needs of trademark owners working in their own languages and scripts.

6. Regional clearinghouses are in conformity with ICANN’s mandate for promoting competition.

Dangers/Problems with the IRT Clearinghouse Proposal

· ICANN has neither scope nor authority to create a single, monolithic, undefined superstructure to adjudicate legal rights. The creation of such a legal regime falls completely outside the function of ICANN as set out in the foundational White Paper.

· Even if ICANN had the authority – which it does not – it should not want to create a single, monolithic IP Clearinghouse because such a process will change the nature of the community ICANN serves. The jockeying for position in the IP Clearinghouse, the hope of extending its reach to even the dimmest trademark and other intellectual property right, will lead to a run on ICANN by trademark owners. This discussion, debate and jockeying for rules of the IP Clearinghouse should happen outside the bounds of ICANN.

· Under its current structure, the Clearinghouse will be open to abuse. On page 13 of the recommendation, the IRT states: “It can also perform similar functions for other types of RPMs besides those recommended by the IRT at this point”. This provides too much discretionary power to the administrator of the Clearinghouse, which can easily lead to abuse.

· There are too many gaps with the current IPC proposal, which should be anticipated that they will increase costs (administrative and external), lead to litigation and create bureaucracy and confusion.

· The IRT report allows any trademark – both registered and unregistered – to be listed. This is wrong – any geographical name, personal name, ‘reading book’ club name will be able to be listed. (Indirect Horizontal Abuse)

· Inclusion in the Clearinghouse of any rights that do not constitute valid federal or national trademarks will require an intense validation process. This is not cost-effective for either ICANN or the registries/registrars.

· The validation concerning established legal rights falls outside the scope of ICANN.

· It is unclear from the recommendation the kind of affidavits trademark owners are expected to submit to the Clearinghouse. There is no mention on verification processes on the validity of these affidavits.

· The Clearinghouse is meant to reduce costs and be efficient. Under this proposal it fails, since it suggests a structure that depends much on ICANN, registries and registrars. Similarly, unfamiliar ICANN practices, combined with non-uniform registry rules, will impair the efficiency of the IPC.

We propose the Private, Regional Trademark Clearinghouses (TMCs) meet the following standards:

· TMCs initially hold data relating to the valid trademark registration issued by a federal or national jurisdiction that conducts substantive examination of trademark applications prior to registration.

· TMCs should not accept data in respect of trademark registrations that relate to what we call ‘domain name services’. The USPTO[1] and the European Union[2] have both declared the invalidity of such registrations.

· The data should be submitted by trademark owners directly to the relevant TMC.

· ICANN foster a TMC in each region. The TMC be encouraged to provide detailed information about the trademark, including the category of goods and services in which the mark is registered. ICANN might further foster a common database search mechanisms for registrars to search trademarks by country, date of registration, international classification number (where applicable), etc. (following the WTO model for trade agreements, http://www.wto.org/english/news_e/pres09_e/pr548_e.htm).

· Trademark owners should submit to the TMC as much relevant information as possible, which should include, but not limited to, proof of valid trademark registration(s) as authorized by National Trademark Offices.

· TMCs shall be operated by neutral service providers that are not currently in a direct contractual relationship with ICANN;

· TMCs must be technically state-of-the-art and their daily operation must enhance the rapid provisioning of domain registrations.

· TMCs must be scalable. For example, they should be able to accommodate identical trademarks registered under different classes of goods or services or in different trademark registration offices, recognizing the territorial nature of trademark law and international classification systems.

· TMCs must be able to accommodate all types of registered trademarks, including word marks and device (logo) marks that contain a word element from every trademark registration office in the world.

· TMCs must be able to deliver fast, accurate information in a standard format using a state-of-art technical platform that is secure and robust.

· TMCs must possess knowledge and expertise in the field of trademark law, processing efficiency, project management skills and risk management skills.

· Listing in the TMC should be optional. It is recommended for registrars and registries to work with the TMCs appropriate to their languages, regions, scripts and community.

· TMCs must also be capable of holding data in the original language and not merely a translation, accommodating this way the particularities of IDNs.

Note: We suggest as a model for future Trademark Clearinghouses the market-models of today, including PriceWaterhouseCoopers for the <.eu> sunrise period. Information can be found: https://www.maxnet.eu/conditions/en/final_presentation_road_shows.pdf

Kathryne Kleiman, Esq.

Trademark Law Attorney, UDRP Drafter

Dr. Konstantinos Komaitis,

Law professor,

Law School,

University of Strathclyde, Glasgow, UK



[1] See Paragraph II.E of USPTO Examination Guide 02-99, http://www.uspto.gov/go/tac/notices/guide299.htm

[2]http://oami.europa.eu/ows/rw/resource/documents/CTM/legalReferences/partb_examination.pdf

The IRT Proposals of a Globally Protected Marks List (GPML): Dangers/Problems And Innovative Solutions

Paper shared on August 7, 2009

Introduction

We spent a long time trying to understand what the IRT wants in the GPML. So, let’s start with how the IRT envisions the GPML – “the IRT recommends the creation of a Globally Protected Marks List to provide protection to Globally Protected Marks at the top and second levels”. [IRT report, p.16)

What the IRT specifically proposes:

Trademark owners that wish to have a mark included on the GPML must provide to the Clearinghouse documented evidence that is capable of being verified of the criteria listed below. After the initial gTLD application round, these criteria should be evaluated and, if appropriate, revised.

The IRT no longer provides the ICANN community with any specific recommended criteria about registration numbers in the various regions, and appears to be at odds within the Committee as to the answers.

Further, all trademark registrations must have issued on or before the date that GPML applications are first accepted and must be based on trademark registration applications filed on or before 1 November, 2008, and the second level domain name for the GPM’s principal online presence must be identical to the GPM. [IRT Recommendation, pp. 16‐7]

Dangers/Problems with the GPML Proposal:

1. Globally Protected Marks (“GPM”): The IRT report seeks to create an alternative, new category of trademarks that do not fall within the famous/well-known category.

2. GPMs will not necessarily involve famous marks, and may not even include famous brands – however, the IRT’s GPMs could include very esoteric scientific and technical terms that are currently used worldwide.

3. Thus, these new GPML marks will be even broader in their numbers and have protection far beyond that of today's well-known brand names.

4. The new GPML marks will have a ‘supernova’ status, and for the discussion below, we call them 'supernova trademarks.'

5. The protection of these supernova marks will be excessive within ICANN.

6. The scope of protection of these supernova marks outside of ICANN is unknown, and potentially quite dangerous.

Note: as an academic specializing in the UDRP and studying its untended consequences outside of ICANN, Dr. Komaitis shares his expert opinion:

"Unfortunately, we cannot tell you much about the scope of protection for ‘supernova’ trademarks, how courts view them, how academics understand them – they don’t exist. Even in the case of famous/well‐known marks, which welcome an additional layer of protection compared to ones not classed as famous, the fame of the mark does not grant its owner immunity against any other use. The GPML will grant this sort of immunity – it will allow owners of marks to exclude words and monopolize the domain name vocabulary.

"The IRT has touched upon (and is asking ICANN to canonize) an issue, which is more controversial than the completely baffling, unchartered and confusing path of famous marks. In the US, as in most parts of the world, one cannot tell it is a famous mark unless the court says so. At an international level there is no single list, no consensus and no agreement on that very issue."

In brief:

· The GPML will exorcise certain words from the DNS as it promotes protection of strings of characters, rather than protection of the mark and its association with goods or services;

· The GPML will create an alternative, new category of trademarks that do not fall within the famous/well-known category;

· The GPML will include very esoteric scientific and technical terms that are currently used worldwide

· The GPML will create ‘supernova’ marks.

· The GPML will create scope of protection of the supernova marks beyond that of even what globally famous brands should enjoy within ICANN;

· The GPML overrides the fundamental legal principle that only courts can determine whether a mark qualifies as “famous”;

· The GPML will help a trademark owners elevate the status of their trademarks to the protection of "text strings," not trademarks -- truly unprecedented levels;

· The GPML will not further consumer protection;

· And the GPML will violate a fundamental norm of Free Speech and Freedom of Expression -- the prohibition on its ‘prior restraint’


Additional Dangers/ Problems for ICANN of Overseeing a GPML

· Harassment - trademark owners will lobby ICANN incessantly for the GPML bars to be set to a point where their company's trademark qualify

· Liability - trademark owners not listed in the GPML may sue ICANN for inclusion for and seek damages for the costs of not being included with the special privileges (envisioned and unforeseen) that the GPML will deliver

· Defend ICANN from potential lawsuits from trademark owners losing their domain name due to GPML hijacking (abuse of a granted GPML listing)

· Defend ICANN from potential lawsuits from free speech advocates

· Process appeals related to the GPML;

· Administravia – e.g. processing addition/deletion from the GPML at regular intervals, including performing assessments for GPML listing candidates.

Conclusion

The issue of globally famous marks is one which has wracked the brains of those who created ICANN, those who worked within ICANN since its inception, and today. When faced with the question of famous marks, the White Paper turned to WIPO.

When faced with the question of famous marks protection in the gTLDs, Working Group B of the DNSO also turned to WIPO -- requesting a list from which to base its findings.

All have turned outside of ICANN -- for its not within the technical scope and mission to create such a list. ICANN must wait until the global community has created standards, and a list, of globally famous marks.

We join NCUC, ALAC, eNom, and many others, even within the IRT, who say that the GPML is clearly a proposal of the IRT that cannot go forward and cannot be fixed.

We hope to make you see that implementation of such a proposal could backfire and fragment the registration of domain names. The degree of protection that trademark owners seek through the GPML does not exist in the offline world.

Kathryn Kleiman, Esq.

Trademark Law Attorney, UDRP Drafter

Dr. Konstantinos Komaitis,

Law Professor,

The Law School,

University of Strathclyde, Galsgow, UK

The IRT Proposals of URS Dangers/Problems And Innovative Solutions Of this Major Battlefield Issue

Paper shared on August 7, 2009

What the IRT Proposes:

The IRT proposes a system to take away domain names which it hopes will:

l monetize off the value and goodwill of a brand,

l distribute counterfeit goods,

l malware and other malicious software,

l phishing attempts, and

l adult content (IRT Final Report, page 25).

The IRT Committee states that its intent is to “supplement and not replace the UDRP,” and “address cases of abusive use of trademarks where there is no genuine contestable issue as to the infringing or abusive use (i.e., not a fair use or commentary situation nor a situation involving questions of whether the registrant is or is not authorized or selling, for examples, legitimate, non-counterfeit goods). (IRT Report, p. 25)

Yet, their proposal is far broader than the types of conduct they actually seek to limit.

Dangers/Problems with the URS Proposal:

In its Joint Statement on the IRT Report from members of ALAC and NCUC, the two groups jointly presented the following severe critique of the URS:

“We have serious issues with the Uniform Rapid Suspension Service (URS)

as proposed. For instance, the URS mechanism subverts conventional

UDRP practice as it gives entirely insufficient time for notice to the

registrant of the pending dispute. Thus, the registrant is unfairly limited

in his/her right of response and the process is missing the fundamental

principle of due process.”

In brief, the URS , if implemented, will:

  1. Replace the UDRP. Faster and cheaper, and without anything to restrict its use to the truly egregious cases, all trademark owners will use the URS, only turning to the UDRP if they lose the URS.
  2. Provide fundamentally unfair notice and response time. It assumes email never goes into SPAM filters, the people never take vacation, and that lawyers never need time to prepare a response. The procedural rights of domain name registrant will, thus, be severely impaired.
  3. Game the URS process entirely in favor of the trademark owner lawyers. The URS games and manipulates the system completely in favor of the trademark owner – whose attorneys can spend infinite time and money on preparing the filing, and then choose, for instance, a filing date in mid-August or mid-December.
  4. Presume the guilt of all respondents, all domain name registrants. “Nowhere in the Benchmark checklist or in the URS is consideration given to the 'proper' respondent. Every respondent is presumed to be guilty upon presentation of the complaint and any expense incurred is foisted upon the respondent without rationale [sic] or comment.” (Keating written comments, p. 12). Further, incomplete answers, even if unintended or inadvertent, regardless of how complete, are deemed a default. The rights of registrants should never be disregarded.
  5. Suppose a clear standard where none exists; disregard rigid standards of legal evaluation. Simply put, the cybersquatting question is not as easy as the IRT Committee would suggest. The “we know it when we see it” approach by trademark owners is disproved by UDRP and court decisions:

· Strick Corporation v James B. Strickland Jr.: “Nothing in trademark law requires that title to domain names that incorporate trademarks or portions of trademarks be provided to trademark holders. To hold otherwise would create an immediate and indefinite monopoly to all famous mark holders on the Internet, by which they would lay claim to all .com domain names, which are arguable ‘the same’ as their marks. The Court may not create such property rights in gross as a matter of dilution law. Trademark law does not support such a monopoly”.

· Banco de Chile S.A. v. Eric S. Bord, Esq: “The clear statement of Section 4(a)(iii) is that the Complainant must prove both bad faith registration and bad faith use. This distinction was not casually made by the UDRP drafters. It contrasts with the terms of US Anticybersquatting Consumer Protection Act (which was being enacted at the same time as the UDRP), which allows a claim based on either bad faith use or bad faith registration. Twisting this very clear requirement on its head, the majority says “non-use” can equal “use” […] Here, there is no evidence of any specific intent to prevent the mark owner from reflecting its mark in a corresponding domain name. To deprive evidence of intent from the simple fact of registration is to ignore the requirement of finding intent. To apply 4(b)(ii) in cases where, as here, the mark owner already has numerous domains corresponding to its mark is illogical. The mark owner has no right to very possible variation on its theme […] it has a right to use its mark in some corresponding domain name, not all of them. Again, cybersquatting is an act that shows particular harm to the mark owner, not some general disadvantage”.

Further, Paul Keating's written comments strongly show, from a registrant attorney perspective, that the IRT fails to provide any quick or clear test for the egregious conduct it seeks to prevent:

Rapid Solution?

The need to rapidly take down clearly abusive sites is clear. However, the URS suspension will not occur until after the ultimate decision. No deadline is imposed on the decision. Given the complete issues that must be addressed (with a genuine issue of material fact exists) it is unrealistic that the remedy will be any faster than the average default UDRP. It is not a 5-minute oil change solution.” (Keating, IRT Comments, p. 9).

  1. The so-called crisis of cybersquatting does not exist. One element of consensus in Sydney, among the entire assembled non-IRT community, was that overall UDRP cases, relative to overall domain name registration, were going down.

  1. We cannot stress enough the disastrous consequences of moving forward with the URS as drafted. If ICANN moves forward with the URS, it will:

- displace a system created by a multi-stakeholder community with one created by only one community (trademark community);

- disregard 10 years of work and experience in the UDRP;

- endorse a very flawed litigation system created from scratch by one biased community;

- displace and bypass the GNSO on a serious policy issue;

- miss the input of those registrant attorneys and academics, who should be involved in drafting such an important dispute proceeding;

- send certain registrant groups to Congress, and

- destroy the GNSO process.

Adoption of the URS “as is” threatens ICANN and its grassroots based system, will lead to explosion within ICANN, and unfair and unjust decisions which will haunt ICANN for years to come.

What We Recommend:

For the URS, we support two recommendations. The first is quick, simple and clean and provided by registrant attorney Paul Keating.

  1. Don't adopt the URS -- Do Modify the UDRP for new gTLDs in a quick, clear way to simply incorporate the goals of the URS – the Keating Answer.

Paul Keating, longtime registrant attorney, states: “I am entirely opposed to the URS. I feel that the primary concern of the TM holders (getting sites shut down quickly and at an efficient cost) can be handled entirely as follows:

1. Amend the UDRP to provide that if a UDRP is filed, the respondent has a period of time to elect any of the following:

a. Defend

b. Surrender the name.

“Electing to defend requires a nominal payment by the REGISTRANT. This acts only to keep the webpage operating pending the final UDRP decision. Failure to file an election to defend does not equate to a default. AND a registrant who did not file an election to defend can come back at any time during the UDRP process and file one, pay the fee and get the domain back up and running. This protects the registrant who may not have seen the notice.” (Keating, Email to Kleiman & Komaitis, 8/5/2009).

Keating's position is reasonable and accurate. His recommended “quick fix” to the UDRP, if approved by the GNSO, would be consistent with GNSO's past work, in keeping with UDRP history, and further, fulfill the needs of the IRT as set in the IRT report. It also would resolve the deep substantive and procedural concerns of the dozens of individuals, businesses and organizations who opposed the URS in the public consultations and written comments.

Keating's solution is a reasonable place to stop and a solution that should be broadly acceptable to the ICANN community.

The second proposal is not as simple, but provides a middle ground. It was proposed by eNom's Richard Tindal in NYC and London and provides a narrowly-tailored focus to the URS that seems to a) satisfy the immediate needs of the IRT Community, and b) not threaten the legitimate rights and interests of good faith registrants by pulling them into a fundamentally unfair process:

II. Preventing the Conduct the IRT Really Intends: Egregious Conduct

(A Possible Middle Ground)

The IRT Committee was very clear in its meeting with NCUC representatives in Sydney: their intent in the URS is to block and/or quickly take down the serial cybersquatter. In this context, the report of the IRT makes much more sense. Serial cybersquatters, especially those who register dozens or even hundreds of variations of well-known marks as domain names to monetize their value with pay per click websites

- are not engaged in fair use;

- are not engaged in a commentary situation; and

- are not in a situation involving questions about whether the registrant is or is not authorized or selling, for example, legitimate, non-counterfeit goods. (All examples of situations the IRT does not intend the URS to encompass). (IRT Report, p. 25).

Therefore, the recommendation, appropriately, of eNom, is to focus on this conduct. If ICANN wants to take steps to prevent serial cybersquatting, then the URS should be narrowed down to address those concerns.

As eNom stated, ICANN can do that by requiring the URS filing to have:

A. A minimum threshold of 26 domain names for a single trademark (with the trademark owner allowed to show that various entities are really controlled or operated by the same party);

B. A minimum threshold of 72 names for multiple trademarks held by the same trademark owner (e.g., the 400 trademarks held by Time Warner) (with the trademark owner allowed to show that various entities are really controlled or operated by the same party); and

C. Domain names identical or confusingly similar to a mark in which the Complainant holds a valid trademark registration for the word mark issued by a jurisdiction that conducts substantive examination of trademark applications prior to registration.

(The “word mark” threshold is critical as design marks do not provide trademark owners with rights to the use of the word outside of the colors, graphics and pictures for which the trademark is sought.)

In this case, the threshold itself, if met, goes a long way to showing the bad faith and bad conduct of the registrant. It is a rapid takedown process for serial cybersquatting – and a direct and innovative solution to the problems raised by the IRT.

A Further Step: Rapid Takedown of Criminal Conduct

We note that the other egregious conduct raised by the IRT Committee in the IRT Report is not solely trademark infringement – it is criminal conduct in its own right. To “distribute counterfeit goods,” to provide “malware and other malicious software,” to engage in stealing PIN numbers and other “phishing attempts” and to post “adult content” under a children's product name to lure underage visitors to a pornography site is criminal conduct!

In such instances, however, trademark issues are secondary -- it is the illegal content of the website itself that creates the cause of action by which the conduct against the domain name and website may most rapidly proceed.

The rapid takedown of domain names involved in criminal matters is the current work of the Anti-Phishing Working Group (APWG). They have procedures in place today, and are formalizing the more extensive Abusive Domain Name Resolution Suspension Process.

Further, every registrar and registry we talked with has “Rapid Takedown Procedures” for criminal conduct. It is a problem with a known solution – and one for which the whole UDRP does not have to be modified or set aside. These are not exclusively or even substantively trademark infringement issues, thus ICANN’s approach to this issue can be narrow, specific and targeted to the serial cybersquatting cases.

Recommended Next Steps:

We strongly recommend that ICANN quickly convene a multistakeholder group to:

- review the two options; and

- determine which option to move forward on.

Should the group choose to move forward with the narrowly-tailored URS, we recommend that ICANN move to quickly convene a UDRP Review process through the GNSO. The IRT discussion has shown that there is much to discuss and improve in the UDRP and 10 years of experience on which to base the discussions. Prior to the delegation of any new gTLDs, the GNSO could be well on its way to reshaping a fuller, better-informed and more tightly drafted UDRP.

Kathryne Kleiman, Esq.

Trademark Law Attorney, UDRP Drafter

Dr. Konstantinos Komaitis

Law Professor

University of Strathclyde,

Glasgow, UK