- Copyright 2016
Domain names are alphanumeric identifiers for websites, and they are exclusive in nature: no two website owners can have identical domain names.1 A domain name comes in the form of alphanumeric text strings which follow the two forward-slashes in a World Wide Web address.2 The domain name system came about as a result of trial-and-error in the 1970s and 1980s when the need arose to direct information from one point to another.3 This system involves two addresses: a computer-readable set of numbers (or Internet Protocol addresses), and a paired name which can be read easily by humans (called the domain name).4 The Domain Name System is organised around a root and a tree structure, divided into hierarchical levels from Top-Level Domains, Second-Level Domains, and so on, up to Nth-Level Domains, i.e., the name of the host computer.5
Since its inception, the domain name system has had an interface with trademark law. Trademarks are distinguishing marks which set apart one’s product, service or brand from others in the same industry. It is not difficult to see why the function that a trademark carries out could easily be considered analogous to that of a domain name online – it is also a unique identifier, which sets apart the online interface of the one who holds that domain name from that of another entity online. Thus, there has been an ongoing battle relating to trademark laws as manifested through domain names in the arena of the Internet as well.
This paper explores two of the mechanisms through which disputes pertaining to trademarks and domain names are resolved. The first is the Uniform Domain Name Dispute Resolution (hereinafter “UDRP”) evolved by the International Corporation for Assigned Names and Numbers (hereinafter “ICANN”)—an alternate dispute resolution mechanism to resolve trademark related disputes in the event that the holder of a trademark feels that a domain-name holder is infringing on his rights to his detriment. The second is the recently evolved Trademark Clearinghouse policy that has been created in the light of the announcement by ICANN in 2012 that it is seeking to expand the number of generic Top Level Domains. The main aim of this paper is to describe these complicated processes, and understand how exactly to achieve that fragile balance between the rights of trademark holders and those of the expression and freedom of the domain name holders in the global Internet ecosystem.
The first section examines the nuances of trademark-related issues surrounding the domain name system. The next section analyses the UDRP, encompassing the rationale behind its origins, the contents of the policy, and the critiques of the system as it stands. An assessment will then be made of ICANN’s new gTLD programme as the source point for the evolution of the Trademark Clearinghouse, along with an analysis of the procedure as well as the various apprehensions that have been articulated with respect to this set-up. The paper concludes with an analysis of the best way forward in both these regards.
Trademark Law and Domain Names
It was in the 1990s when a private company named Network Solutions Inc. (hereinafter “NSI”) was given the monopoly for registering domain names on the internet; thus began the privatisation—and commercialisation—of the internet.6 Needless to say, as individuals began registering domain names, claims were made over the domain-name space, inevitably leading to trademark law complications. Domain names are analogous to property in the digital space. Being unique identifiers, a single domain name can be held only by one person or entity at one time.7 It has been said that trademark law views the internet as a single marketplace.8 While trademark law allows two people in different geographical or product markets to hold the same trademark simultaneously, any given domain name can only be held by one person, and moreover, this domain name will be global and not restricted to a single geographical space, and are outside the scope of domestic legal systems.9 As per the Lanham Act of 1946, to prove a trademark infringement, one must be able to demonstrate: (a) a protectable interest in a valid trademark; (b) defendant’s commercial use of that mark; and (c) a likelihood of confusion as a result of such use.10
While domain names do not have the same value as a trademark, there are numerous attempts made by companies to utilise their trademarks as domain names as this trademark is the name which the average customer is likely to search for the company online.11 This is threatened by the problem of cybersquatting, which WIPO views as one of the most significant threats to the holders of trademarks.12 The site of most trademark issues and consumer confusion related to domain names is the second level domain name, as that is what consumers use to locate the website, and arrive at an understanding of the business of the company.13
‘Cybersquatting’ is a term used to refer to the deliberate, bad-faith, abusive registration of domain names, with the speculative intent of reserving trademarks as internet domain names with the mala fide purpose of selling them to the trademark owners at a much higher price.14 A corollary of this is ‘cyberpiracy’, where there is a violation of copyright, as the registrant uses a confusing domain name to mislead customers to their website for commercial gain.
An interesting high-profile case of cybersquatting occurred in 1995, when American entrepreneur, Dennis Toeppen registered 200 domain names corresponding both with generic terms as well as with trademarks of the times, going on to state that domain names were, “…valuable underdeveloped real estate…(with) absolutely no statutory or case law regarding trademarks in the context of Internet domain names at the time.”15 At that time, the only option available for complainants who were obviously disgruntled by this turn of events was to approach the court in trademark-related litigation.16 Toeppen was eventually sued and found guilty of trademark dilution, and the court opined that cybersquatting was an effective threat to the proper functioning of trademarks.17
Traditionally, trademark infringement is based on the notion of there being a likelihood of confusion in the minds of the consumers as a result of the alleged infringing act.18 This is easily applied to cyberpiracy cases; however, the issue of cybersquatting will not always fall under this definition, as the registration of the domain name does not cause confusion in the minds of customers. Regardless, though, it is violative of the rights of the holder of the trademark. Most of the time, such a cybersquatter will not post anything on the website, implying that she would have used the mark, which is required to be proven in cases of infringement.19
Trademark dilution is another avenue by which the mark owners have traditionally sought to bring action against the rights holders.20 In the US, the Federal Trademark Dilution Act provided for injunction in the event that another person’s commercial use of a trademark caused dilution of the distinctive quality of the trademark.21 This kind of dilution could imply blurring – i.e., the unauthorised use of a mark for dissimilar goods that may reduce the value of the mark as a unique identifier22 – or tarnishment, i.e., the association of a mark with an inferior quality such that its reputation is harmed.23 However, while most cases of cybersquatting may fall under these categories,24 in some instances, the act of cybersquatting may lead to neither, as seen in the case of Panavision v. Toeppen,25 where the defendant merely registered the trademarked domain name and sat on with the intention of selling to the defendant without seeking to post anything. The court based its decision on the harm that was caused as the Panavision mark’s capacity to act as a distinguishing factor for its goods and services on the internet had been reduced. The implication of this kind of logic was that the likelihood of the trademark holder winning in case of ‘famous’ marks would be higher, while the rights of those who held marks that were not famous might not be adequately protected, as the dilution would be harder to prove, or, as seen in the Hasbro26 case, the courts might extend broadly the definition of what amounted to a famous mark.
In 1999, the US Congress passed the Anticybersquatting Consumer Protection Act, creating civil liability for cybersquatting, and expanded the reach of the existing law beyond ‘famous’ marks, but limited commercial use to instances where there was a bad-faith intent.27 However, fair-use exceptions were allowed in this regard.
It is pertinent to note that few other jurisdictions have a law specifically to deal with cybersquatting – the UK and Germany, for instance, resolve cybersquatting matters using trademark law.28 A landmark case in this regard is Harrods v. UK Network Service Ltd.29 as per which the court accepted that the law related to trademark and passing off could extend to domain names.
Uniform Domain Name Dispute Resolution Policy
The UDRP is an international dispute resolution procedure that is under the mandate of ICANN, which allows trademark holders to challenge the registrant of an Internet domain name, and solve the issue by agreement, court action or arbitration.30 As Froomkin has analysed, the power of the UDRP stems from ICANN’s solo control over one of the most critical internet resources, viz., domain names.31 One can only acquire domain names from a registrar with the right to allocate names in an ICANN-accredited domain name registry.32 It is the authority of ICANN which decides which registries are authoritative.
This policy came up in 1999, after the United States Department of Commerce gave ICANN the authority to control domain name registrations.33 Prior to this, the NSI had its own dispute resolution policy, created in 1995, according to which the complainant had to give notice to the registrant that there had been a violation of trademark, following which the registrant would have to attempt to prove that she owned a trademark in the contested name, failing which the NSI would put the domain name on hold till resolution was reached between parties, either through agreement or through litigation.34 This was obviously problematic as the NSI did not actually provide a process for the resolution of disputes in the first place. It was also argued that this policy facilitated reverse domain name hijacking through the ease by which the alleged infringer’s domain name could be placed on hold, and the limited defence available to the registrant consisting solely of showing that she owned the trademark in the domain name in question.35
In 1999, after the inception of ICANN, the NTIA, through a White Paper, called for a study on domain names and trademark issues from the WIPO, inviting the WIPO to “…(1) develop recommendations for a uniform approach to resolving trademark/domain name disputes involving cyberpiracy (as opposed to conflicts between trademark holders with legitimate competing rights), (2) recommend a process for protecting famous trademarks in the generic top level domains, and (3) evaluate the effects, based on studies conducted by independent organizations.”36 In this White Paper, the NTIA acknowledged that the rights of trademark holders were getting affected, while at the same time conceding that there were also other legitimate rights to counter these that needed protection.37 WIPO recommended that the scope of the new policy be restricted to bad faith, abusive domain name registrations or cybersquatting, and not be extended to disputes between two parties with legitimate competing interests, and that it be able to ameliorate the use of court litigation to resolve such processes.38 In September 1999, the draft of the implementation documents for this policy were prepared, and after public comment, ICANN adopted a second set of implementation documents, resulting in the UDRP, adopted in January 2000.39
The UDRP is a policy that has been adopted by all ICANN-accredited registrars, and it is applicable between the registrar and the customer.40 Paragraph 4(a) outlines when the domain-name holder is required to submit to a mandatory administrative proceeding, in the event that there is a complaint made such that (i) the domain name is identical or confusingly similar to a trademark or service mark in which the complainant has rights; (ii) the domain name holder has no rights or legitimate interests in respect of the domain name; and (iii) the domain name has been registered and is being used in bad faith by the holder of said name.41 All these three elements must be present in the complaint. “Bad faith” as used in Paragraph 4(a)(iii) has been defined in the policy to cover any of four instances – first, where the circumstances indicate that the domain name was acquired for the sole purpose of transferring to the complainant (the legitimate owner of the trademark) for excessive consideration (i.e., not in cases where the consideration covered reasonable costs); second, where the registration was to prevent the owner of the mark from reflecting the mark in a corresponding domain name; third, where it was registered to disrupt a competitor’s business; and fourth, where it was registered to create a likelihood of confusion to attract internet users towards the website for commercial gain.42 This is a non-exclusive list, and arbitrators have added to the definition.
ICANN provides for three affirmative defences for the domain-name holder in whose name the domain has been registered.43 One protects those who can demonstrate that they intended to use the domain name in connection to a bona fide offering of goods or services prior to them receiving the notice.44 The second defence is that the holder of the domain name has acquired recognition by the domain name despite him lacking a trademark. 45The third is that the domain name holder is using the name legitimately for non-commercial purposes, or his use qualifies as fair use without any intent for commercial gain to mislead consumers.46 This drew from US Trademark law at the period when the UDRP was introduced, when a purely non-commercial use was seen as a strong defense against an infringement claim; however, it drew a limit around this by providing for the tarnishment concept as a restriction on such non-commercial use.47
The policy creates the institution of the “Provider”, i.e., an administrative dispute resolution provider, and gives the complainant the power to select the provider from amongst those approved by ICANN.48 Currently, there are five such providers – the Asian Domain Name Dispute Resolution Centre; the National Arbitration Forum; WIPO; the Czech Arbitration Court Arbitration Center for Internet Disputes; and the Arab Center for Domain Name Dispute Resolution.49 The provider selected then goes on to administer the proceeding, appointing a panel for the same.50 The proceedings are guided by the Rules of Procedure.51
In terms of costs, the fees are to be borne by the complainant, unless the domain name holder chooses to expand the administrative panel in size from one to three.52 ICANN itself refrains from participation in the proceedings.53 This proceeding, most importantly, does not bar the jurisdiction of the courts.54
Another persistent criticism of the UDRP is its costs. While the expenditure is certainly less than that of litigation, especially under the US Court system, it nonetheless costs the defendant anything between $5000-$10000, and possibly more as the matter is pursued.55 Given that there is a significant chunk of defaults, there is a tendency for meritless complaints to be filed, for which the brunt is often borne by innocent respondents who do not meet the bad-faith description. This is unethical, and the loophole needs to be repaired.
Most arbitration is not precedent-based, and the UDRP is no exception to that rule; however, in many cases under the UDRP, precedent has been cited and used to arrive at a reasoning behind the cases.56 Mueller goes on to state that most of these cases that have set a precedent have resulted in an expansive interpretation of the rights of the trademark holder, and the definition of ‘bad faith’ has been inflated.57 Telstra v. Nuclear Marshmallows,58 for instance, has served as a precedent for the principle that even the passive holding of a domain name constitutes bad-faith usage (even if there is no actual use). There is also a case which goes to state that it is bad faith if the respondent is not contributing any value to the internet59 which has been frequently cited, which stretches the definition to breaking point. Thus, there seems to be excessive power vested upon the panellists deciding UDRP disputes, which is problematic as they have demonstrated arbitrary decision making, which often results in de facto precedent being set. Another related criticism is that the result of the UDRP dispute is heavily contingent on the panellists who are chosen to arbitrate the case, as well as the Provider which is used.60 For instance, the WIPO and the NAF are perceived widely as being biased in favour of complainants, and there is therefore evidence of extensive forum shopping by complainants – in 2014, for instance, WIPO decided 2634 UDRP cases, and NAF decided 1557, of which complainants prevailed a massive 92 percent of the time if there was a default (which was true for about 1000 of the cases), while when the respondent did respond, the success rate for them was about 46 percent.61 This could possibly be because of the composition of these panels. WIPO panels are largely composed of trademark lawyers and professors, while NAF is constituted largely by retired US justices.61 It has also been seen that there is a dramatic difference in outcomes when decided by three-member panels vis-à-vis single-member panels – the win rate by complainants was about 83 percent in 2001 for single-member panels, as compared to 60 percent in three-member panels.63
The Trademark Clearinghouse was conceptualised in 2012, when ICANN announced that it was going to introduce a new gTLD programme to expand the number of generic Top-Level Domains beyond the existing 21 known ones.64 This new policy allows anyone, for a fee, to register any new top level domain.65 ICANN had introduced this policy as a means to protect trademark owner’s rights to prevent overreliance on .com, and to allow trademark holders to apply for . protections to create a secure online location for their brand.66 ICANN envisioned this as reducing the problem of cybersquatting and subsequent dilution of the brand.
The Trademark Clearinghouse has been described as “…the most important rights protection mechanism built into ICANN’s new gTLD programme.”67 It is a centralised database of verified trademarks, connected to the new gTLDs that ICANN with launch.68 Once the trademark holder places his or her mark in the Clearinghouse, s/he can register the trademark as a second-level domain under a newly delegated gTLD before the registry opens the gTLD to the general public, and once a legitimate trademark holder places his/her mark in the Clearinghouse, no one else can obtain the second-level domain.69 For instance, for the gTLD .icecream, if Haagen Dazs chooses to place their mark in the Clearinghouse, they would get priority over the domain name haagendazs.icecream by virtue of their trademark. However, if there was another company by the same name, they would not have the same priority rights. It was conceptualised in 2009 by a team of 18 IP specialists who were entrusted with the task of producing a report to find solutions for potential trademark risks stemming from the new gTLD expansion programme.70 One of their most important recommendations was the creation of an Intellectual Property Clearinghouse, which they felt would adequately protect the existing rights of trademark owners. The clearinghouse would be sustainable, and benefit registries, users as well as consumers.71 They outlined the following roles for the IP Clearinghouse:72
Some of their more important recommendations were that the IP Clearinghouse be maintained by a neutral service provider, be able to accommodate identical trademarks registered under different classes of goods and geographical locations given the territorial nature of trademark law, be able also to accommodate all types of registered trademarks (Words, logos, and others).73 Finally, it recommended that the IP Clearinghouse provide the following services74 – (a) annual validation of trademark rights, (b) a Globally Protected Marks List, (c) A Pre-Launch IP Claims Service76 and (d) data for taking part in Uniform Rapid Suspension mechanisms before the registration of the trademark.77
Based on these recommendations and public comments, ICANN launched the TMCH policy in March 2013.78
The procedure for this has been outlined in the Applicant Guidebook issued by ICANN. To acquire a new gTLD, an applicant is to submit a fee of $185,000 and indicate whether it wants to apply for a .brand gTLD (say .mamagoto), an industry gTLD (say .basil), a geographic gTLD (say .Digboi) or a non-Latin script gTLD (say .मैंदिखाऊं).79 The application must also be marked as either being ‘open’ or ‘community-based’.80 The former gives the applicant the right to restrict the gTLD, exclude competition and/or restrict it. The latter implies that the applicant must use the gTLD for the benefit of the community in question.81 If the application then passes the evaluation stage, the applicant must mandatorily enter into an agreement with ICANN to provide trademark protection through a mandatory Trademark Claims Service and Sunrise Service Policy for a minimum 30 days before the launch—this enables trademark holders to protect against use of their mark by precluding third parties from registering the identical domain under that gTLD.82 This is where the role of the Trademark Clearinghouse becomes important. The Trademark Clearinghouse is not a rights protection mechanism unto itself – it is essentially a database which provides information to the new gTLD registries for supporting pre-launch Sunrise or Claims Services.83 Word marks that can be included in the Clearinghouse include nationally or regionally registered marks, marks validated through judicial proceedings, and marks protected by a statute or treaty in effect at the time the information is submitted into the Clearinghouse as well as other marks constituting intellectual property”.84
The Sunrise Service Policy allows trademark holders the opportunity to register domain names related to their marks before the names are available to the public, a period of about 30 days.85 If a given TLD (say .icecream) is applicable to a certain good or service, the owner of a trademark in this industry (say Kwality) would be likely to want to ensure that its mark is registered under the TLD, and the Sunrise period facilitates that in an expeditious way. The Trademark Claims service lasts for about 90 days following the initial registration, and if, during this period, there is an attempt to register a domain name matching a mark recorded in the Trademark Clearinghouse, the person making the attempt will receive a notification; if they still go ahead and register the domain name, the TMCH will send a notice to the trademark holder, who can then request the registry to take action to resolve the issue immediately.86 However, this remains the limited extent of rights available under this mechanism. While they do not inherently prevent infringement, if registration proceeds even after receipt of notice, it is easier to prove bad faith. It is probably anticipated that rights holders use TMCH in conjunction with the UDRP or the URS. Moreover, if the holder of the trademark has previously won a UDRP or a court decision on the trademark it is filing with the TMCH, the trademark owner can elect to file a verified trademark and 50 names similar to that trademark under the “Abused Doman Name Label Service.”87 TMCH validates the URS procedure, as recommended in the 2009 report discussed earlier.
It is important to note that the TMCH exists merely to validate the data, and does not determine the substance or scope of the rights. Two identical trademarks from different jurisdictions can therefore coexist under this scheme.
A major concern that has been raised is that the TMCH could eventually end up challenging the traditional system of trademark registrations, despite it being a voluntary and optional service.88 Despite that, ICANN has been vested with the authority to maintain an important database, and this vests them with plenty of discretionary power; some critics are concerned that the TMCH might end up substituting national trademark offices.89 This seems to be a little far-fetched as of now, given that inclusion into the TMCH is not an indication of anything but the validity of an existing trademark. This inclusion itself is based on existing trademarks vested by national trademark registries, and non-inclusion into the TMCH does not invalidate the trademark. No additional trademark rights are created through inclusion into the TMCH.
That said, it cannot be denied that the process of inclusion is costly, and is likely to favour those with larger brand presence and financial clout. There may also be significant confusion in this respect. For instance, the gTLD .microsoft may not create as much trademark confusion, given that ‘Microsoft’ is a fairly well known registered trademark. However, the gTLD .apple can lead to multiple SLD contentions, given that it is a generic word as well as one of the largest brands in the world. There is a real worry that given the clout and presence commanded by a large and established brand such as Apple Computers, Inc., their rights may be overwhelmingly prioritised over, say, the trademark held by a local apple selling establishment in a third world country. Further, say in the case of two identical trademarks in different industries, it would be unclear to the consumer which was being referred to – for instance, .guardian could refer to the insurance company “Guardian” or the news media house “Guardian”. The likelihood of confusion increases in this respect, especially with .brand trademarks. The Sunrise Policy, it has been argued, regards the DNS as flat namespace, and fosters consumer confusion, as well as allows parties the space to block multiple second-level domain names without necessarily using them, therefore promoting empty tracts of namespace.90
The TMCH Service provider has been given the discretion of determining what additional trademark-like information it wishes to include.91 There have also been privacy concerns raised about the TMCH Data.92 These could be resolved through the development of strong policy regarding the same – discussions are ongoing even now on RPMs and their reform.
Some people raise concerns about the need for the URS in the first place, arguing that the UDRP seems to be functioning effectively, and that the URS policy is heavily biased against the defendant.93
It is seen that while the UDRP policy was created with the intention of reducing the burden of litigation, and as a system for ensuring fairness for all rights holders, it has somehow evolved and morphed into a system which accords strong privileges to the rights of the trademark holders. Even though the definitions within the policy seem fairly narrow, the ways in which the UDRP panels have adjudicated have expanded the meaning of ‘bad-faith’ under the policy far beyond its original mandate. It must be kept in mind that the founding rationale behind the creation of the UDRP was to check the offence of malicious cybersquatting. This seems to have somehow given way to an expensive system which is unfavourable to the domain name holder even if he or she has absolutely no mala-fide intention of breaching trademark. The uniqueness of the domain name system does not seem to be guiding the adjudicators of the panels, and any suggestions at reforms must begin with a revamp of the constitution of the panels. Instead of focusing on how the trademark is used, there is a tendency to focus on if the trademark is used at all, and if this is coupled with default on the part of the respondent, his rights are not taken into account most of the time. This is an area within the UDRP that needs reform, and urgently, so that more equitable decisions are brought out to serve the original purpose of this policy.
With respect to the TMCH, it is early days yet. The major areas of concern here are with respect to brand confusion and the hegemony of well established brands over rightful identical trademark holders. It is important to keep in mind that while the Clearinghouse itself is not a dispute resolution system, it has bearings on whose rights are going to be prioritised, especially under the URS policy. Some have suggested that the URS policy be suspended altogether, given that it is so heavily biased towards the rights of the trademark holder, even more so than the UDRP. There may be some merit to that, as the existing UDRP already provides for a process, and with reform, ought to suffice for the purpose.
The purpose of these policies is not to foster monopolies over the Internet, a resource which has since its inception been based on principles of freedom, ease of access, and multistakeholderism. It is therefore imperative that the way forward find a balance between the interests of all the players in this ecosystem, and not give undue privilege to those with financial clout and brand value to make their presence felt.
PRIVACY & DATA PROTECTION
A specialist neurology clinic in Japan has this revealing signage with its website address prominently proclaiming its function (photo courtesy:...
Abstract Critical information infrastructure (CII) is a pillar on which modern nations function. The revolution in information and communication technologies...
No longer the subject of science fiction, Artificial Intelligence (AI) is profoundly transforming our daily lives. While computers have been...
Only eight years after India passed the Information Technology Act, did the term cybersecurity appear in a statute through a...
The creation of the Internet Governance Forum (IGF) was a watershed moment in the history of the Internet. This article...