INTERNET DRAFT O. Crepin-Leblond Global Information Highway Ltd December 1996 Expires: June 1997 Delegation of International Top Level Domains (iTLDs) draft-iahc-crepinleblond-itld-prop01.txt Status of this Memo This document is an Internet-Draft. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts. Internet-Drafts are draft documents valid for a maximum of six months. Internet-Drafts may be updated, replaced, or obsoleted by other documents at any time. It is not appropriate to use Internet- Drafts as reference material or to cite them other than as a working draft or work in progress. To learn the current status of any Internet-Draft, please check the 1id-abstracts.txt listing contained in the Internet-Drafts Shadow Directories on ds.internic.net, nic.nordu.net, ftp.nisc.sri.com, or munnari.oz.au. 0. Abstract This document comments on: - policies and procedures to: o allow open competition in domain name registration in the iTLDs, o allow multiple registries for the COM and other iTLDs, - proposals to create new international top level domains (iTLDs), - the danger of partinioning the Internet if exclusive iTLDs were allocated. 1. Introduction For the purpose of delegation, the current top level domains (TLDs) fall into the categories listed below: o National TLDs such as UK, FR, SE, ... ZW are named in accordance with ISO 3166 (except UK). In the major part, those TLDs are allocated to national naming registries which are either held by that country's telecommunication monopoly, or an organisation of providers, or a government's ministry. Delegation of these TLDs is undertaken by the Internet Assigned Number Authority (IANA) according to RFC1591. Discussion of the suitability or unsuitability of such top level domains is outside the scope of this document; o US Governmental TLDs (GOV, MIL) restricted to US government related entities. Discussion of the suitability or unsuitability of such top level domains is outside the scope of this document; o Specialised TLDs (for example INT) which are closely controlled by IANA. Discussion of the suitability or unsuitability of such top level domains is outside scope of this document; o Educational Institutions TLD (EDU) restricted to educational institutions such as universities, schools, colleges, and educational consortia - recently the registrations in this category have been restricted to 4 year colleges and universities. Discussion of the suitability or unsuitability of such top level domains is outside the scope of this document; o Shared International TLDs (NET, COM, ORG) are usable throughout the world with few restrictions (NET for computers of net providers etc.). Clearly, in the above list, the Top Level Domains which are considered as being over-crowded are the Shared International TLDs, and especially the COM domain, designed for commercial entities. The reasons for this over-crowding of domain space are numerous but mostly come down to the fact that most US companies are registered under the COM TLD, while at the same time, non-US companies are also allowed to register under that domain. This international TLD is therefore suitable, and desirable for any company or commercial organisation that wishes to have or portray an international presence. The end of the monopoly on assignment of domains within the above TLDs by Network Solutions, Inc. (NSI) has prompted the writing of a memo by Jon Postel, from the Internet Assigned Numbers Authority (IANA) on October 1996 entitled "New Registries and iTLDs". Part of the work described in this document is outlined in an Internet Draft by Bush et. al, January 1996, entitled "Delegation of iTLDs" and constructive input by other people mentioned in Postel's memo. It is on this memo and on other proposals that the current draft of this paper will comment on. 2. Goals The goals of this memo are similar to those described in Section 2 of Postel's "New Registries and iTLDs" October 1996 document. Namely: "To facilitate administration of the domain name subsystem within the Internet by ensuring that there is an open and competitive marketplace for clients to obtain and subsequently maintain delegation of sub- domains within the iTLDs, while preserving the operational integrity of the Internet DNS itself." Preservation of the operational integrity of the Internet DNS itself is a prime concern. The document referenced above opens the door to other proposals within the IAHC discussion taking place. The ultimate goal is to ensure the safe transition of the DNS naming process from being run by a sole central monopolistic registry to a new system of registries that will ensure cost-effective and efficient registration processes, while ensuring the Internet's future development. Avoiding pitfalls of creating a recipe for future failure and anarchy within the DNS is an issue to be addressed while it is still time to do so. 3.0 Scope of this Document The goal of this document is to critically examine the arguments put forward by Jon Postel's October 1996 "New Registries and iTLDs" memo, as well as other IAHC November and December 1996 draft proposals regarding the re-organisation of Top Level Domain space. This draft wants to specifically: 3.1 Argue that "trademark dispute avoidance" as being the reason for requirement of brand new international TLDs to be created is in fact an invalid reason. The creation of new iTLDs will in no way help in trademark dispute resolution. We argue that it will probably make the matters worse with trademarks being disputed in more than one top level domain. 3.2 Argue that "provision of directory services" as being the reason for requirement of brand new international TLDs to be created is in fact an invalid reason. The creation of top level domains following classification of activity is invalid due to language difficulties and the provision of directory services needs to be undertaken as a completely separate service. The DNS is a way to name computers around the world, not a directory enquiry service, and should be kept as is. 3.3 Argue that broadening the top level namespace to allow "simple" and "easy to remember" names for users will in fact make the DNS top level domains confusing to general users if more than 10 new top level domains are created at most. 3.4 Argue that the creation of exclusive top level domains will create a string of problems, including the creation of unstable top level domains, locking-up of clients in a particular registry, and may trigger the downfall of the Internet as a whole due to the psychological partitioning aspect of namespace this will create. 3.5 Re-iterate points which were present in the Bush January 1996 "Delegation of iTLDs" Internet Draft, which are out of the picture now because that document has expired on 22 July 1996. 3.6 Propose the creation of a limited number of international Top Level Domains, starting the process with the creation of a single new international Top Level Domain to conduct the experiment of multiple registries / multiple domains one level down as proposed in Section 3.5 above. 4. Technical Assumptions All technical assumptions are the same as Jon Postel's October 1996 "New Registries and iTLDs" memo. 5. Creation of iTLDs for trademark dispute avoidance There are no international trademarks. Both sides of the story about trademarks are summarised in Postel's Appendix A.1.1 Trademark Issues. I would however suggest that while the number of not so strong trademark owners far outweighs the number of strong trademark owners, it is not up to the Internet DNS to find a solution to the problem of similar trademarks in different fields. Not so strong trademark owners from different countries can make use of their country top level domain to differentiate between each other. Not so strong trademark owners from the same country can extend their domain name. For example, for a company FOO proposing computer services, and FOO being in the bakery business, the following are proposed: foo-computers.com / foocomputers.com / foo_computers.com and foo-bakers.com / foobakers.com / foo_bakers.com The creation of additional International TLDs such as .BAKERS and .COMPUTERS is serving the same purpose while restricting the range of activities. This restriction will severely hinder the flexibility of domain names according to the field of activity of the company using that top level domain. The result is directly opposite to that described in the goals (section 2) of this memo. This should be avoided. 6. Provision of Directory Services using the DNS It has been argued that new iTLDs should be created to provide for a framework capable of supporting directory services. A proposed classification of domains takes into account the field of activity of a company and it is then possible for that company to register its domain under the international TLD fitting their activity. This would result in a straight forward way of looking up domain names of companies, by putting a structure "per activity" to the DNS. The creation of top level domains following classification of activity is invalid due to language difficulties. Proposed categories take only account of the English language. To be effective, the categories should match similar categories in the world's most commonly used languages. The few hundred English language categories will translate to several thousand categories, each one triggering the creation of another international Top Level Domain. The result will be an unworkable DNS, with several thousands of iTLDs, most of which will not be understood by users because they are in a foreign language. This will serve to confuse users and qualify the Internet as being much too complex for everyday use. The aim is to make things easier for the majority, and creating several thousands of iTLDs will not achieve this. The idea of using latin or easily understandable acronyms as iTLD is another proposition which would solve the language problem. The trouble is that while it was easy to find the current iTLDs (COM, ORG, GOV, etc.) which are understood easily in most common languages, it is impossible to find several hundreds of iTLDs which are understood easily. Latin is not a solution. The provision of directory services should be undertaken as a completely separate service. Significant resources should be devoted to that task which requires constant updating of non- internet-related information. Any attempt to let a directory create itself spontaneously is doomed to failure through lack of interest when a significant amount of time has to be "donated" to that task without rewards. Furthermore, companies should have a choice on whether they want to be included in the directory or not. 7. Creation of "simple" and "easy to remember" company names The argument against adding new international Top Level Domains to broaden the domain hierarchy in order to create simple and easy to remember company names is the same as that put forward in Section 5 of this memo regarding Trademark dispute resolution. Namely: companies from different countries can make use of their country top level domain to differentiate between each other. Companies from the same country can extend their domain name. Restrictions on company names and activities have also been addressed in Section 5 of this memo. Furthermore, as the Internet expands, it will be increasingly difficult to find a company's name / address without making use of a professional directory of companies. While today most Internet users resort to blindly trying the name of a company bounded by "www" and "com" in order to access their Web site, the chances of success will diminish as more and more companies enforce their presence on the Internet. Users will shift their search patterns by first trying a top level directory of companies as described in Section 6 of this memo. In addition, the creation of any more than 10 new international Top Level Domains is likely to confuse the average Internet user. Currently, users can somehow make use of the Top Level Domain and have a broad idea of the type of correspondent reached, whether on a country level, or type of activity (Commercial, Internet Service Provider, US Government, Non-profit, etc.). Adding a significant number of more international TLDs will complicate the equation. 8. Creation of Exclusive Top Level Domains There have been proposals for the creation of a significant number of International Top Level Domains on a "per registry basis". This effectively means that a registry is chartered to have complete control over a specific iTLD. One of the aims of the exercise of re-organising the registration system once the monopoly of Network Solutions Inc. expires is to organise an open and competitive marketplace for clients to obtain and subsequently maintain delegation of subdomains within the iTLDs. Delegation of iTLDs on an exclusive basis will only serve to create a monopoly for each exclusive iTLD. Furthermore, a client "locked" into a registry's exclusive iTLD will not be allowed to migrate to another registry if it is not satisfied with the service from that registry, unless it gives up the domain within the iTLD. Effectively a client will have to change address when changing registry, except if it registers itself in a common, shared domain, such as the ones already in use (COM, ORG, etc.). In view of this fact, international Top Level Domains allocated exclusively to a single registry look like very bad news for a client. Since the Internet is set to attract more clients than the total number of users already connected, it is safe to assume that the majority of new domain name allocations will take place with newcomers to the Internet, who have no knowledge in what Top Level Domain to choose. Companies will be pushed by registries to apply for the registry's exclusive TLD. The result is likely to be a lot of bad press for the Internet through the exposing of "make.money.fast" rogue operators of some registries. An additional danger is the "orphaning" of some international Top Level Domains if the registry having exclusive rights to that iTLD were to go bust or lose its license to operate. A special case of allocation of an exclusive iTLD is possible when a large company or internet connectivity provider registers all of its hosts under that exclusive iTLD. Such a practice is likely to enhance the provider's psychological standing in the mind of prospective clients by making the company sound "grand". Having one's own iTLD could be seen as a status symbol for some of the larger connectivity providers. Such a practice will be detrimental to the Internet on the whole: it may induce the psychological partitioning of the namespace, thus endangering the Internet's integrity. No longer will the Internet be seen as "one worldwide network", but we will go back to defining separate networks, each with its own set of rules, gateways, iTLD etc. 9. Re-iteration of a point in a previous Internet Draft document A very valid point was mentioned in the Internet Draft of Bush et al. "Delegation of iTLDs", January 1996, Appendix A: "It has been suggested that the root domain be opened up to a fairly arbitrary registration of new iTLDs. While the idea has its social appeal, o it would be an irreversible decision with no prior experience and o some argue that TLDs, like global variables, should be few and created with great care. Therefore we suggest that one of the early public service proposals that should be seriously considered would be one which proposes a shared iTLD which would have very open creation of sub-registries; thereby conducting the 1,000 domain experiment one level down." 10. Proposal I propose that a limited number of international Top Level Domains be created for specific purposes, that number being not significantly larger than 10. Furthermore, I propose going ahead with the suggestion described in Section 9 of this memo, namely: The creation of a shared iTLD in which the public service proposals of all sorts could be carried out, as a means to "test the system" without endangering the whole structure of the DNS and limiting any damage if things go wrong. Conducting the experiment one level down will limit the risk of serious damage, and is the natural cautious approach adopted in any project the size of the one attempted by introducing radical changes to a world-wide system such as the Internet. Any re-organisation having the potential of being a "recipe for catastrophe" should be avoided. A non cautious approach is a "recipe for catastrophe". 11. Security Considerations There are no known security considerations beyond those already existant in the DNS. 12. Main References In this memo, the following Internet drafts are often quoted: Jon Postel (IANA), "New Registries and the Delegation of International Top Level Domains", October 1996. Randy Bush (RGnet), Brian E. Carpenter (CERN), Jon Postel (ISI), "Delegation of International Top Level Domains (iTLDs)", January 1996. 13. Authors' Address Olivier MJ Crepin-Leblond Managing Director Global Information Highway Limited Phone: +33 (0)956 841113 8 Palace Place Mansions Fax: +33 (0)171 937 7666 36 Kensington Court Email: ocl@gih.com London W8 5BB United Kingdom