New generic top-level domains (gTLDs) are heading this way whether we relish the prospect or not. After several years of discussion amongst the internet community, at its meeting in July in Yokohama ICANN adopted a policy for the introduction of new gTLDs in a "measured and responsible manner".

It is largely, although not universally, accepted that .com is congested. It is of course true that combinations of words, invented words and misspelt words are still available, but this does not assist the established company who cannot simply have its name as a .com. Many companies, especially those who woke up to the e-commerce revolution rather late in the morning, discovered that the most obvious domain name or names for their company were already taken by other companies operating in different fields or different countries. This squeeze on the most popular TLDs has created a demand that is currently being partly met by the new style country code domains (ccTLDs) which market themselves on a global level, although often focussing on a particular sector (.tv, .md, .nu etc). The popularity of these "global" ccTLDs could be viewed as further evidence for the need for new gTLDs.

Although the long term plan is to introduce dozens of new gTLDs, ICANN has determined that they should be introduced slowly. There are likely to be 5-6 (or even less) new gTLDs initially. There is some debate as to whether these should be truly generic, similar to .com, or whether they should be "chartered" ie relevant to a specific sector (.travel, .movie, .museum etc). It is likely that most of the new gTLDs will be "chartered" and this should be welcomed as a way of making the internet more user friendly to its ever increasing number of users, provided there is a reliable method of ensuring that websites operating within the chartered gTLD do operate in the relevant sector.

Formal applications from those seeking to sponsor or operate one or more of the new gTLDs will be accepted by ICANN from 5 September to 2 October . However, before you rush to submit suggestions, a $50,000 non-refundable fee is payable for the submission of an application to become a sponsor or operator of a registry. $50,000 may deter some, but there is a huge amount of money to be made by having the rights to register a potentially popular new domain name, such as .biz, which will attract those who missed out on .com, or .sex, which would undoubtedly be a popular TLD (and make the task of screening material unsuitable for children easier).

The date for completion of the negotiations with the new sponsors and operators is the end of this year. This is likely to coincide with the introduction of another new TLD, one which will fall somewhere between the traditional ccTLDs and the gTLDs, that is .eu. The European Commission has adopted a timetable of consultation which is currently likely to result in the introduction of .eu early next year.

One issue which has delayed the introduction of the new gTLDs is the concern by the intellectual property community that new domains will inevitably lead to new cybersquatting problems. Large companies which have faced hundreds and in some cases thousands of cybersquatting problems with the existing domains do not relish the idea of having to defend their brands against cybersquatters in the new gTLDs.

Famous Marks

Initially there was a proposal to establish protection for famous trade marks. It is famous trade marks which suffer from cybersquatters, and the theory went that if a list of famous marks could be prepared, these marks could be exclusively reserved for the owners. However, attempting to put the theory into practice proved problematical. Registrars were reluctant to have to screen applications, which would increase their costs and lead to delays. There was no consensus as to the extent of protection which should be offered. If "Disney" was accepted as a famous mark, should this just prevent registration of disney.TLD by anyone other than the Disney Corporation, or should it also protect against combinations which included the mark eg disneyfilms.TLD, 1disney.TLD, ilovedisney.TLD or disneysucks.TLD?

Added to this was the difficulty of proving fame. Although the idea of well-known marks is not new (it is enshrined in the Paris Convention), a system for establishing fame on a global basis has never been established. Other than a few truly global marks, such as "Coca-Cola", many marks which are famous on a national or regional basis, cannot claim to be famous in all parts of the world. As the internet is a truly global forum, this led to great difficulty in formulating an appropriate test to establish global fame. It then started to become apparent that even if a workable test could be drafted, many companies may be reluctant to submit their marks for examination. In many jurisdictions particular rights attract to marks which can claim to be well-known or famous. If a mark had been submitted for evaluation of global fame and had failed, it could prejudice its national protection. Although the tests were likely to be different, the fear of having an opponent raise the question of failure to be acknowledged as a famous mark, especially before a US jury, led some companies to back away from the proposed "famous mark list".

The Sunrise +20 Proposal

Acknowledging the difficulties inherent in a system which relied on establishing fame, the Intellectual Property Constituency (IPC) came up with a new proposal which would not require any such assessment. Famous marks are invariably registered as trade marks. The "Sunrise +20 Proposal" was therefore based on registered trade marks, without the need to prove fame. It was proposed that registered trade mark owners be given 30 days prior to the opening of the new TLDs to the public, to register their trade marks along with 20 variations on that trade mark. It was argued that this merely reflected the protection given to trade marks under existing law. However, trade marks in some parts of the world are not examined for distinctiveness and marks can be registered without ever having been used. The trade mark system protects marks in certain distinct classes of goods and services and does not give universal rights to the owner to prevent another trader from using the mark. This is why in the UK the mark "Lloyds" for example, is used and registered by different companies in relation to pharmacy services, banking services and insurance services. To provide the owner of a registered trade mark with the right to register its mark as a new TLD ahead of a company with that name who does not have a registered trade mark, or ahead of a company who has built up a reputation by trading under that name, would give an unfair and unjustified advantage to trade mark owners.

The Names Council failed to support the Sunrise +20 Proposal earlier this year.

Daybreak Proposal

The IPC then suggested the "Daybreak Proposal" as an alternative to "Sunrise +20". The Daybreak Proposal would operate in the same way, but would allow trade mark owners prior registration in the new commercial TLDs with only the exact or precise trade mark, as registered.


It is generally agreed that all new gTLDs should be subject to the Uniform Dispute Resolution Procedure (UDRP), which gives trade mark and other rights owners recourse against cybersquatters. ICANN has resolved that in introducing the new gTLDs it will give heed to the "importance of appropriate protections of rights of others, including intellectual property rights, in connection with the operation of the TLD, especially during the start-up phases". It is possible that this will mean adopting a scheme along the lines of the Daybreak Proposal or it may mean ensuring that the UDRP is available to those whose rights are infringed by cybersquatters.

The UDRP is generally working well for trade mark owners and although it puts the cost of protection back onto the trade mark owner, who has to pay $750+ to bring proceedings, which cannot be recovered against the cybersquatter, it does present an easy way for ICANN to be seen to be paying heed to the needs of the intellectual property community, without burdening the registrars with any form of screening process. The UDRP could also be enhanced to provide a forum for resolving disputes concerning the "chartered" domains.

It is now up to the applicants seeking to win the ICANN Board of Directors' approval to run Registries relating to the new gTLDs to propose mechanisms that will adequately protect intellectual property rights and they will have to make the judgment call as to how much protection will be sufficient to satisfy the Board, without committing themselves to an overly complex system which would effect their own profitability.

In the future as more and more gTLDs are introduced the value of a domain name (and therefore the value in cybersquatting) is likely to reduce. If there are 200 possible and 20 likely TLDs which a company could have, consumers are not going to methodically type in all the possible combinations in the hope of eventually finding the website they are seeking. More sophisticated ways will have to be established to help users find what they are looking for. In the meantime, the race for the new gTLDs and the likely disputes which will follow will their role out are only a few months away.

First published in ILR in September 2000.