Legal Briefing Paper from Law Professor Jacqueline Lipton on GNSO Recommendations for Domain Name Policy

[GNSO] recommendations contemplate that ICANN should implement a process that would accommodate the introduction of new gTLDs that are not currently available to Internet domain name registrants or registries. In its recommendation paper, it contemplates several principles for deciding on strings of characters that may be utilized in a new gTLD. These principles include:

• New strings should not infringe the existing legal rights of others (Recommendation 3).

• New strings should not be contrary to generally accepted legal norms relating to morality/public order (Recommendation 6).

• Applications for new strings should be rejected or deferred if there is substantial opposition to a relevant string from ‘among significant established institutions of the economic sector, or cultural or language community, to which it is targeted or which it is intended to support’ (Recommendation 8).

• ICANN staff evaluators will make preliminary determinations about applications for new gTLD strings (Recommendation 11).

The NCUC and IP Justice have raised particular concerns about aspects of these recommendations. Specifically, they are concerned that some of ICANN’s proposals give too much weight to trademark holders’ interests without giving sufficient weight to other competing legal interests in words and phrases, such as those arising from legal concepts of free speech. They have also voiced concerns that under Recommendation 6, ICANN may by default be trying to legislate internationally for morality and public order and that this may not be an appropriate burden for ICANN, as opposed to national lawmakers. They raise related concerns with respect to ICANN Recommendations 8 and 11 in the sense that these recommendations focus more on international legal and cultural norms than on the technical capacities and functions of ICANN. Recommendation 8 also raises the specter of censorship in the introduction/use of new gTLDs by bodies that have not been clearly defined in the ICANN proposals. It is also unclear how decisions would be made as to the rejection or deferral of new strings on this basis. Which organizations would be consulted? Whose policies would be applied? What experts, if any, would ICANN consult?

ICANN GNSO Recommendations 5, 9 and 12

I would add some similar concerns about the following ICANN recommendations:

• New strings should not include country, territory or place names or words describing countries, territories, languages or peoples in the absence of agreement with relevant governments or public authorities (Recommendation 5).

• Applications for new gTLDs must entail a clear and pre-published application process using ‘objective’ and ‘measurable’ criteria (Recommendation 9).

• Dispute resolution processes must be established prior to the start of the relevant process (Recommendation 12).

Recommendation 5. This recommendation raises the specter of government censorship or control of particular gTLDs. This may or may not prove to be a problem in practice given the existence of two character country-code top level domains (ccTLDs) within the current system. These ccTLDs have apparently not, to date, created major problems, at least as compared with some of the issues arising under currently available gTLDs. However, it is possible that a new gTLD string pertaining to a country would prove to be more desirable than a corresponding ccTLD and this recommendation may give imbalances of power or control over particular new gTLDs to certain governments or public authorities. In some ways this concern mirrors the concerns of IP Justice and the NCUC about Recommendation 8 to the extent that it is unclear under that recommendation whose policies should be protected in the decision to defer or reject registration of a particular gTLD string. An associated concern with recommendation 5 is that it may not always be clear who is the relevant government or public authority who would need to agree to the use of a particular new gTLD: for example, would all Asian countries have to agree to the use of a ‘.asia’ gTLD and, if so, how should ‘Asian country’ be defined in this context and who should define it? Moreoever, who should decide which ‘public authorities’ should be consulted about use of particular new gTLDs? How should ‘public authority’ be defined here?

Recommendation 9. This recommendation calls for the use of pre-published ‘objective’ and ‘measurable’ criteria in the application process for new gTLDs. It is not clear how ICANN per se would establish such criteria. If it is contemplated that ICANN would consult relevant national and international bodies or individuals in discharging this problem, then perhaps this recommendation is not so problematic. However, such a consultation process would likely take a long time and may slow down the introduction of new gTLDs for a considerable period. Such a process would entail: (a) identifying relevant expert bodies; (b) consulting with them on relevant issues: and, (c) translating relevant issues into a set of pre-published objective and measurable criteria for the new gTLD application process. This further assumes that such issues are indeed transferable to objective and measurable criteria.

Recommendation 12. Dispute resolution processes may be much more problematic in practice than contemplated by ICANN’s recommendation 12. My assumption is that Recommendation 12 refers to simple dispute resolution processes for new gTLDs such as those currently in effect under the Uniform Domain Name Dispute Resolution Policy (UDRP) for some existing gTLDs. The problem here is that dispute resolution processes that take account of multiple legal interests outside commercial trademark interests are not easy in practice. Different jurisdictions, and different bodies within the same jurisdiction, may diverge widely in attitudes and even in laws on free speech, public order etc. Arbitrators under simple UDRP-style dispute resolution processes may not be equipped to handle these kinds of disputes. Dispute resolution procedures may therefore have to be somewhat more complex than is currently contemplated by ICANN if they are to take account of a variety of competing legal interests, rather than merely trademark interests. For example, while there are some things a simple arbitration process can handle well, there are other things that are much more complex and difficult and may need to be turned over to national courts or experts.

General Discussion

It is important to start re-focusing the regulation of the Internet domain name system generally on interests outside of pure trademark interests. The introduction of new gTLDs and the development of processes for introducing them may provide a good opportunity for achieving this goal. However, any attempt to regulate broad policy issues relating to social and cultural norms on speech, public order and morality in domain names will be very difficult for any national or international body or group. ICANN also faces the practical difficulty that its major area of expertise is technical and functional. It is therefore important for ICANN to clarify what groups, bodies or individuals it might utilize in carrying out future legal and social developments within development of its domain name processes. In particular, ICANN should consider more specifically who to consult in formalizing specific processes for: (a) the introduction of new gTLD strings; (b) establishing dispute resolution procedures for those strings; and, (c) deciding whether the introduction of particular new strings should be deferred or rejected.

It should also be noted at the outset that many of the key problems identified by ICANN, IP Justice and the NCUC reflect legal issues that have arisen in the past with respect to existing gTLDs, although perhaps in slightly different contexts. In other words, the balance between trademark interests and other legitimate interests in Internet domain names, for example, has already proved problematic in situations involving disputes about registration and use of domain names under existing gTLDs (notably .com, .org and .net). Thus, in many ways, the ‘balance of interests’ questions in the new gTLD debates could be regarded as an extension of unresolved issues under current domain name laws and policies. The addition of new gTLD processes will likely exacerbate existing legal problems. The upside of this is that it may, and hopefully will, provide a new forum for addressing some of these problems.

In my view, it is important to put the debate about new gTLD processes into its historical context in order to properly address the concerns that have been raised here. So please bear with me for a couple of paragraphs while I describe this context and why it is important now. The current framework for regulating disputes relating to ‘.com’, ‘.net’ and ‘.org’ domain names has been focused largely on the protection of commercial trademark holders against cybersquatters. There is little harmonized attention paid to the protection of other legitimate interests in relevant Internet domain names within this framework. This is evidenced in the drafting of the UDRP and the American Anti-Cybersquatting Consumer Protection Act (ACPA). While these regulations do make allowances for ‘legitimate interests’ in domain names where relevant domain names have not be registered or used in bad faith, they do not set out rules to affirmatively protect non-trademark-based registrations and uses of .com,. .org or .net domain names. This is not particularly surprising because it was not the intention behind these rules to do so.

The historical focus on the protection of trademarks against bad faith cybersquatters is understandable within its context. These were key concerns of relevant regulators in the mid to late 1990s when e-commerce was in its infancy, and governments wanted to encourage this new medium of commerce. It was widely thought – although not universally agreed – that bad faith cybersquatting per se was a socially wasteful activity that potentially harmed the development of electronic commerce without producing any associated social benefits. There is probably nothing inherently wrong with the UDRP and ACPA in this respect. They did deal with a real world problem and, in many respects, they are now old news. Presumably, this is why debates today about the introduction of new gTLD processes do not dwell on the rules and regulations implemented in 1999. However, those rules and regulations have raised new post-1999 problems that have not yet been addressed in a systematic way.

Issues under the existing domain name system that relate to the balance of trademark interests with other legitimate interests in domain names do include the need to balance trademark interests with interests such as: interests in personal names, cultural and geographic indicators, free speech (including the right to parody, comment on and criticize a trademark holder), other basic human rights, and rights to free and democratic government. ICANN has identified some of these issues in its recommendations. IP Justice and the NCUC have raised concerns about clarification of, as well as appropriate implementation of, ICANN’s stated goals here.

The main problem for ICANN in identifying and implementing these kinds of ‘interest balancing ideals’ is that, as with its administration of existing gTLDs, ICANN’s expertise is largely technical and functional. It is not a body staffed with people whose main expertise is to deal with these difficult balances of competing legal and social interests in multiple societies around the world. Effectively bringing debates about international public order and morality, as well as free speech and human rights generally, into a predominantly technical process comes at a high cost. However, failing to address these issues in a relevant forum also comes at a high cost, as previous and current experiences have shown us.

What is needed at this point is a combination of the following: (1) ensuring that the technical aspects of this process do not somehow become a default proxy to legislate for important and complex national and international social, cultural and legal norms; (2) more clearly identifying bodies or individuals who can appropriately identify and make recommendations on relevant issues within the development of the more technical aspects of the process; and, (3) ensuring that these bodies are brought into the relevant process in time to prevent damage to important legal and social interests. To some extent, that may be what is happening at the moment, but this process may need to be more formalized to avoid exacerbating some of the problems that have arisen in the past under the current domain name system.

Conclusions

The aim of this briefing paper has been to raise awareness of ideas that may be pertinent in the ongoing process to develop new gTLDs. My hope is that this paper generates, or at least facilitates, useful debate in this context. There are, as yet, no clear solutions to many of the problems addressed. We seem to be at a point in the development of the new gTLD processes where it would be useful to more fully identify and discuss relevant legal and social issues, as well as bodies and individuals that may be best suited to advise on them, and ultimately help draft and implement regulations about them where possible. This is an important time in the development of the domain name system and this kind of debate and development would prove extremely useful, particularly in order to avoid some of the practical problems with respect to new gTLDs that are already evident in the administration of domain names registered under existing gTLDs.

Jacqueline D. Lipton, Ph.D.
Professor of Law
Co-Director, Center for Law, Technology and the Arts
Associate Director, Frederick K Cox Center for International Law
Case Western Reserve University School of Law
11075 East Blvd, Cleveland, Ohio, 44106, USA
Email: JDL14@case.edu
Web: http://law.case.edu/faculty/faculty_detail.asp?id=129