Historical Replication Notice This page is a faithful replication of a historical work originally published in 1996 as a website by Jonathan Agmon, Stacey Halpern and David Pauker while LL.M. candidates at The George Washington University Law School. It is reproduced here by Agmon Law Pte. Ltd. for historical and educational purposes. The content reflects the law, technology and internet infrastructure as they existed in 1996 and has not been updated. Nothing on this page constitutes legal advice. Most hyperlinks referenced in the original no longer function.
What's in a Name? (1996) › The Problem and Criteria for a Solution
Introduction

Most people involved with the administration and maintenance of the Internet agree the system for assigning and structuring domain names must be changed. There seems to be little agreement, however, between technicians, academics and commercial interests about that new system.

While there have been several widely divergent proposals, there has not been a clear statement of what must be accomplished by the new system. Content dictates method, and if there is agreement about the goals for the system it will be easier to form a structure.

This page offers criteria for revising the system, and suggestions regarding possible attributes for a revised structure for the domain names themselves.

A. The Problem
  1. Domain names may be indicators of not only location, but also source — and associated with that identified source may be an expectation of quality. While indications of source — with an implied association of quality and reliability — so as to distinguish the goods and services of one person from another, is the classic underlying rationale behind trademarks, the present system inadequately deals with this inherent trademarks nature of domain names.
  2. The current system for obtaining, identifying and administering domain names does not take account of their trademark nature. Anyone can register any name not already taken, and trademark related disputes are increasing dramatically.
  3. Now, because the domain name system provides no reliable indications of source about what may be found at any Internet address, there is commercial uncertainty and consumer wariness. A domain name system which provides confidence regarding the identity of the person, or company, and the source of goods found at a domain name will help make the internet commercially inviting.
  4. Though a domain name has world-wide consequences, there is little or no international coordination regarding either, on the one hand, potential trademark infringement, or, on the other hand, multiple users who may want the same name.
  5. The present system of domain names does not offer enough possibilities to satisfy the foreseeable need. Domain names assigned by NSI — those outside the military, government and education Top Level Domains — are largely one word identifiers. There are, for example, at least 34 businesses in Washington, D.C. alone which begin with the name "Lincoln"; if any one of these used the domain name "lincoln.com" it would prevent anyone else in Washington or anywhere else in the United States from using it.
  6. Disputes about domain name issues — whether for trademark infringement or other grounds — are generally resolved on an ad hoc basis.
  7. In the United States, NSI's Dispute Resolution Policy does not take account of common law or state registered trademarks, unfair business practices, dilution, or conflicts with even well known marks.
  8. NSI's Dispute Resolution Policy is an imposed contract predicated on unequal bargaining power, failing to provide a proper mechanism for adjudicating disputes. NSI, a private company, is acting in a quasi-judicial manner with limited mechanisms for judicial review.
  9. Current proposals for modification of the present domain name system consider elaborate means for administration, but don't define the problem or the goals.
B. Criteria for a Solution
  1. The system for domain names must take into account the international nature of the Internet.
  2. Because there is inherent interaction between trademark principles and domain names, the system for domain names should take the reliability and principles of trademark law into consideration:
    • a. Domain names should have the capacity to be reliable indicators of source — even if the identity of that source is unknown.
    • b. The domain name system should encompass mechanisms to avoid consumer confusion.
    • c. Current trademarks should be accommodated, including "famous" marks, and, in the United States, state and common law trademarks.
  3. Indicators in the domain names should be intuitive for ease of use. Arbitrary division of the "com" domain — for example, into "com1", "com2", etc. — as some have suggested, would cut against this need.
  4. Domain name owners should be able to register logical domain names — that is, those that are related to the identity of the business or person and that are easy to find and figure out.
  5. The system should provide protection against trademark infringement and unfair trade practices.
  6. It should be easy to obtain and maintain domain names.
  7. The cost of obtaining and maintaining domain names should be correlated to the range of use.
  8. The system must be self financing.
  9. Any system must have the capacity to accommodate more than expected future growth.
  10. The system must give certainty and reliability. Administration of the system should be reliable and free of bureaucracy, and any administrative decisions subject to impartial judicial review.
  11. The system must provide for due process.
  12. The system should provide criteria for jurisdiction and conflicts of laws.
  13. Initial dispute resolution should be swift and inexpensive.
  14. The system should have consumer protection provisions.
C. Possible Attributes for a Domain Name System

1. The domain name system should have localized identification

a. There is no territoriality in Cyberspace, but territoriality exists in the Slow-World outside Cyberspace. Location and territorial scope of use in the Slow-World can be important ways to help differentiate the user's identity.

b. The Internet is not only a means of communication and exchange of information, but also a market place for goods. It should be easy for patrons to make dinner reservations at their local Michelle's Brassiere whether in Paris or in Nice, or whether in New York or Los Angeles. The patron should be able to find Michelle's intuitively, and local businesses should be able to have a domain name which identifies them and indicates their Slow-World territorial limitations.

c. Geographic specificity in domain names would help avoid confusion regarding source. It would make words more available for "local" use, and local businesses, for example, could more easily obtain an identifiable domain name of their choice.

d. In the United States, local geographical indicators could be chosen by starting with the top 50 cities or the top 50 Demographic Marketing Areas. But all countries should use the same method for making internal division.

2. Registration would indicate territorial scope of businesses in the Slow-World that are local, regional, national, continental, hemispheric or world-wide

Coca-Cola, McDonald's and Porsche, for example, should be able to register their names on a world basis. On the other hand, it might be more appropriate for Star Television to have a hemispheric or regional registration.

3. Rather than arbitrarily dividing the .com Top Level Domain (TLD), divide it into logical categories of goods, services and personal use

Distinguishing domain names for classes of goods would help make them intuitive and better identifiers. One possibility for a starting point is the International Classifications of Trademark System established by the World Intellectual Property Organization (WIPO) under the terms of the Nice Treaty. Clearly, there would have to be adjustments and expansions to accommodate personal use and to reorient the classifications toward consumers. The suggestion to use this existing International System as a starting point is not an endorsement of the System, but one suggestion of an approach with an expansive beginning.

4. The cost of registering and maintaining a domain name should reflect the number of classes covered and also the scope of territoriality

If Porsche, for example, wants a domain name which includes classes for vehicles, glasses and clothing on a world-wide basis, registration and maintenance fees should reflect that scope.

5. Registration and renewal

a. The initial and renewal terms for registration should be longer than one year. Though, the term for initial registration for commercial domain names might be one or two years since may new businesses fail within the first year or two; terms for renewal should, in any event, be longer. b. Domain names would be periodically renewable provided they are in use.

c. Domain name applications would be published on the Internet allowing for opposition.

d. For an individual or local business, registration and renewal should be inexpensive, and registration should involve no more than a check for conflicts.

6. Registrations within the borders of a country should be handled within that country, coordinating with other national centers or with an international center

An International Center should exclusively handle continental, hemispheric and world registrations.

7. A possible new structure for domain names now within the .com TLD might include the following:

your-name-here.food.nyc.eus

  • 1st Level TLD: This is the current country designator expanded to three letters allowing indications of locality with more specificity — "eus" for Eastern United States. This TLD would also indicate regional, hemispheric or world designations where appropriate — for example, "wld" for a world-wide designation.
  • 2nd Level TLD: This is an additional geographic designator which indicates physical location inside a country — "nyc" for New York City.
  • 3rd Level TLD: This is the current 2nd TLD, including .com and .co, for example. This TLD should be expanded to at least four letters allowing for better and intuitive descriptions of classes.
  • 4th Level TLD: This is the current 3rd TLD — the place for individuals or businesses personal domain indicators should continue to allow up to 22 characters.
D. Need for an International Treaty

Several of the criteria outlined in Section B. can only be achieved by international treaty.

  1. The inevitable growth and use of domain names as trademarks, demands an international convention.
  2. Any national unilateral implementation of a revised domain name system is unsatisfactory because conflict between local and world-wide trademark conventions will inevitably occur.
  3. If the Internet is to grow as a marketplace for goods, services, information, and communication the system for domain names must encompass traditional trademark principles.
  4. A treaty will give businesses the certainty and reliability they need.
  5. An international convention can give consumers protection.
  6. An international domain name system must be guaranteed freedom from interference by arbitrary governmental action.
E. Conclusion

There is widespread awareness of endemic shortcomings in the present domain name system and its administration. These problems should be solved if the Internet is to continue to grow and reach its potential as a populous commercial and information system, for growth demands a system which benefits consumers. If consumers feel comfortable using it, then businesses will prosper, and information will flow more openly.

The Internet is non-territorial and at the same time of world-wide concern. Because solution of the domain names problem demands a universal standard, no single jurisdiction can impose a system on the international community. There is an opportunity to expand ongoing discussions of the Madrid Trademark Convention to include issues regarding the relationship between trademarks and domain names. But, of course, the structure and operation of the Internet itself, though related, should be the subjects of separate and distinct considerations.

Readings about the domain name system and trademarks:

http://aldea.com/cix/maher.html David W. Maher, Trademarks On the Internet: Who's In Charge?
http://www.urich.edu/~jolt/v1i1/burk.html Dan L. Burk, Trademarks Along the Infobahn: A First Look at the Emerging Law of Cybermarks, 1 Richmond Journal of Law & Technology 1 (1995).
http://www.hotwired.com/wired/2.10/departments/electrosphere/mcdonalds.html Joshua Quittner, "Billions Registered," Wired Magazine (Dec. 4, 1995).
http://www.patents.com/nylj1.sht Carl Oppedahl, Internet Domain Names that Infringe Trademarks.
Scroll to Top

Megan Tay

Associate (legal) 

Intellectual Property, Commercial, M&A, Litigation

Bio: Ms. Tay joined Agmon Law after acquiring her experience in a number of international law firms and has completed her training at one of the largest Singapore law firms. She has experience in litigation and intellectual property, and her practices focuses on trademarks prosecution and litigation, as well as commercial law. 

Admissions: Waiting Admission – Member of the Bar of the Supreme Court of the Republic of Singapore since 2026.

Memberships: 

Education: Law LLB from National University of Singapore– 2024

Languages: English, Chinese.

Liangshu Huang

Patent Associate

Intellectual Property, Patents, Litigation Support

Bio: Mr. Huang joined Agmon Law after working in an international patent law firm in Singapore. Mr. Huang assists with patent related drafting and prosecution including in the fields of medical devices, mechanical, biotech, software related inventions, CRI, physics, and defense. Mr. Huang brings to the team the experience and precision required in the field of patents. 

Education: Graduate Certificate in Intellectual Property Law, National University of Singapore (2015-2016)

MRES in Translational Medicine, Imperial College London (2009-2010)

Upper 2nd Class Honors in Biomedical Sciences, University of Bradford (2006-2008)

Languages: English, Chinese and Malay.

Jonathan Agmon

Founder & Director

Intellectual Property, Commercial, M&A, Litigation

Bio: With thirty over years of experience Jonathan specializes in intellectual property, commercial law and litigation. Jonathan established Agmon Law in 2025 after 25 years with Soroker Agmon Nordman, which he co-founded. He is a technologist-turned-lawyer and has represented numerous companies over the years. Aside from prosecution and transactional work, Jonathan also handles contentious matters such as oppositions and cancellations proceedings, and acts both as counsel and arbitrator in litigation, mediations and arbitration. His technology background is in computing and he still codes today having coded some of the Firm’s systems. 

Admissions: Foreign Registered Lawyer (Singapore) – 2016; Attorney and Counselor at Law (New York, United States) – 1996; Advocate, (Israel) – 1995

U.S. District Court, Southern and Northern District of New York; United States Court of Federal Claims.

Memberships: Member of CIArb; Internet Committee at ECTA – European Communities Trademark Association; APAA, INTA (Pro Bono Comittee); Domain Name Panelist (WIPO, FORUM, ADNDRC (HK), ISOC-IL, MFSD)

Education: Master of Laws (LL.M. Intellectual Property), George Washington University Law School – 1996; Bachelor of Laws (LL.B. Hons.), University of East London – 1993; Beijing Languages University: Chinese (Mandarin) – 2009

Languages: English, Hebrew, Chinese-Mandarin (Work in progress), Japanese (working on it…)