Domain Name Comments Received After the August 18, 1997 Deadline

Number: 407
From: Neil Hoener <neil.hoener@fatalerr.com>
To: NTIADC40.NTIAHQ40(dns)
Date: 8/19/97 12:44am
Subject: Registration of Domain Names


Hello Dns@ntia.doc.gov!


I run a 2 line Bbs in Colorado Springs, Colorado and have for more than six
years. I also hold a class C internet domain for the bbs through my ISP and
use an UUCP connection for email for my users. My Bbs, Fatal Error Bbs
(Fatalerr.com), is non-profit and I run it because I enjoy the hobby and the
interaction with the users.


The current arrangement used to register InterNIC domain names is rediculous.
I pay the current InterNIC registar $50 per year for what? To maintain a
server that keeps track of my domain name for the few people that send email
to my very simple operation?


If I am going to register my domain name with someone, let it be a government
agency that has the power to grant a licence and have some sense of public
benefit rather than a private monopoly. Current InterNIC operation is
indicitive of the money-grubbing mind thought that is strangling the Internet
today. Don't just regulate the upper level ISPs, help the smaller domain
names against those that gouge us.


Neil Hoener
Fatal Error Bbs
Colorado Springs, CO
Fido 1:128/103
nhoener@fatalerr.com


* WCE 2.0/2399 * In a VERY weak moment, I started a BBS.


###

Number: 408
From: "Gordon F. B. Johnson" <gjohnson@sysnet.net>
To: NTIADC40.NTIAHQ40(dns)
Date: 8/19/97 7:33pm
Subject: DNS names


The new TLDs seem like a great idea. Things could get a little sticky
if the same host name is available in all the TLDs. And wouldn't a huge
number of places to register names lead to unneccesary trouble from
different people and companies all trying to register the same name at
the same time? Once a name is submitted and online the issue is
settled, but what about a situation where a company is in the process of
getting up and going and its procedure to register its name with one
registrar is delayed long enough for a name bandit (for lack of a
known/better term) to register the name with another registrar and
ransom the name. In general it seems a bit peculiar that a company's
registered trademark name is open to anyone's taking in the internet
naming game.




###

Number: 409
From: Nick Marcou <nick@reston.ans.net>
To: NTIADC40.NTIAHQ40(dns)
Date: 8/19/97 10:39am
Subject: DNS opinion


Make sure it supports 128-bit addressing, for future expansion, but it
should stay with the
x.x.x.x format for IPs, and as small a number as possible of top-level
domains.


Nick Marcou
nick@ans.net




###

Number: 410
From: "Joseph H. Alhadeff" <uscib24@pop.interport.net>
To: NTIADC40.NTIAHQ40(dns)
Date: 8/19/97 4:30pm
Subject: Response to Request for Comments on DNS




August 18, 1997


Via E-Mail
Ms. Patrice Washington
Office of Public Affairs,
National Telecommunications Information Administration
dns@ntia.doc.gov


Dear Ms. Washington:


Please find attached the preliminary comments of the US Council for
International Business (USCIB) in response to the Department of
Commerce's Request for Comments on the Registration and Administration
of Internet Domain Names. These comments are submitted in response to
Domain Name issues which have come to the fore as a result of the
decision not to renew NSI's contract. While commentators have
recognized problems with the current administration of the Internet
Domain Name and Registration system, we are not yet convinced that the
technology underlying alternate proposals has been sufficiently tested
to justify implementation by mid-1998.


In reviewing alternate proposals, our comments focus on the IAHC
proposal which is cited by many commentators as the most likely
alternative to the current system of domain name registration and
administration. In focusing on the issues related to the most
topical proposed Domain Name System solution, we do not address
issues related to alternative solutions based on enhanced numbering
or directory schemes and do not attempt to establish benchmarks for
what constitutes adequate or appropriate protection for trademarks
and tradenames on the Internet.


The USCIB is the American member of the Business and Industry Advisory
Committee to the OECD and the American affiliate of the International
Chamber of Commerce. The USCIB will be preparing further comments on
the Domain Name administration and registration system as part of the
OECD's review of this issue. The USCIB looks forward to continued
work with NTIA and the US Delegation to the OECD on this issue.


Sincerely,




Joseph H. Alhadeff
Director, Electronic Commerce


Attachment USCIBDNS.doc is a Word 6.0 document.


Joseph H. Alhadeff
Director, Electronic Commerce
U.S. Council for International Business
jalhadeff@uscib.org


CC: NTIADC40.SMTP40("Information.Policy.Ad.Hoc.Group@y...

USCIB COMMENTS IN RESPONSE TO THE REQUEST FOR COMMENTS ON THE REGISTRATION AND ADMINISTRATION OF INTERNET DOMAIN NAMES

1. There is no critical impending deadline for the fragmentation of the Internet; getting the right solution is more important than getting a quick solution.

In March of 1998 the NSF's cooperative agreement with NSI to administer three Top Level Domains (.com, .org, .net) will expire. There has been a presumption that the solution to the Domain Name issue must be in place by that date, or shortly thereafter, to avoid some calamity. While we agree that a solution by that date is preferable, we do not believe that the system is likely to stop operating or degrade in any significant manner. The Root Servers and Domain Name Server infrastructure will remain intact on April 1, 1998 and ISP's and companies will continue to follow current usage until such time as an internationally agreed upon solution is adopted and properly implemented.

There is at present both a political and philosophical debate as to the nature of the Internet. Is it an international resource, controlled by none, but accessible to all? Does the US have any special role in light of its founding sponsorship and current dominance? There are also questions of what parties need to be joined in a debate whose results will have far reaching impacts on the administration and operation of the Internet. While an important sociological discussion, this debate clouds the issue of whether a solution needs to be in place by mid-1998.

2. The Concerns of Business.

The USCIB believes that preventing fragmentation of the Internet and maintaining a coherent and interoperable infrastructure are the main concerns of business and should also be the main and immediate concerns of government. The USCIB also recognizes the equally important concerns created by the intellectual property value which has been associated with URLs in the .com domain. Problems of piracy, or cybersquatting, coupled with the lack of effective dispute resolution mechanisms have increased the overhead costs of commerce on the Internet. Lastly, the USCIB recognizes that the inability of two companies with similar names, but substantially dissimilar products, to both operate a "company name".com website has lead to claims of a scarcity of Domain Names, but is also cognizant that this may be an unavoidable technological limitation of the medium.

3. USCIB supports open and interoperable electronic commerce solutions

The USCIB has been a strong proponent of industry-led initiatives to support an open and interoperable electronic commerce marketplace. The USCIB recognizes the benefits of competition in the marketplace as the best way to spur innovation, moderate prices and empower entrepreneurship. Many of the concepts inherent in the IAHC proposal are consistent with these principles.

4. Proposals and potential problems.

The current NSI system has been criticized by some in the US for inadequate dispute resolution mechanisms and is suspect in some international jurisdictions because it is "monopolistic" or "too American". Many proposals concerning the future of the Internet's Domain Name Administration and Registration System have suggested ways of supplementing, improving or replacing the NSI system. NSI itself has proposed a solution which would create competition by allowing the creation of many exclusive domains.

Other proposals include alternative registration schemes, numeric addressing solutions and enhanced directory services, or combinations of these approaches. While a number of these solutions may have merit, most have neither been sufficiently publicized nor vetted to be credible alternatives to the current system of registration and administration. Among the possible solutions which have been more broadly discussed, the IAHC proposal has gained the most attention and may be the current leading contender among the alternatives to the present NSI system.

On its face the IAHC proposal provides a seeming solution to many of the concerns of business in this area, but some of the technological solutions may require further review or testing and the policies may need to be more fleshed out. Many of the other proposed solutions have similar, if not greater, technological and policy issues which need to be addressed. The IAHC proposal is highlighted because of the greater international attention it has received and higher level of momentum it has gained.

The IAHC proposal would end the de facto monopoly on Domain Name Registration and replace it with an unlimited number of registrars. The proposal is based on technological solutions which create secure and scaleable systems with necessary redundancy to provide for a much larger number of technologically less sophisticated players to act as registrars without substantially increasing the risk of technological problems.

The IAHC proposal and initial implementation originally called for 28 registrars. Presumably the original plan foresaw the implementation of technology that would serve 28 registrars with the potential for a gradual increase of registrars over time. This would permit time for implementation, further development and a learning curve for both users and developers to determine how best to proceed. Many members of the USCIB are concerned that the increase of complexity from one to an unlimited number of registrars, with little time for learning curve benefits to be factored in, may increase the risk of technological problems. The concern is not that the technology cannot be built; the software and computer resources already exist to accomplish this task. The concern centers around the need for more time. Time to properly develop, implement and test the system and time to resolve legal and organizational issues which may only arise once the system begins operation.

The roster of IAHC members and supporters reads much like a who's who of the Internet. Their collective reputation provides credibility that entitles their proposal to a hearing - not a presumption of adoption, but a fair hearing. Before a true hearing of the proposal can be had, however, many members of the USCIB are of the opinion that the technology must be completed and tested (for scaleability and interoperability) with results made generally available for review. To our knowledge, no functional testbed has been completed on a small scale, much less a scale that would approximate the number of registrars and volume of registrations contemplated. It should also be pointed out that a few companies have already been formed and are taking orders for registration of company names in the new subdomains at the first possible opportunity. Since many of the piracy problems are obviated by first-in-time registration, there is a possibility that many of the new subdomains will be swamped when first available for registration. The system may be near critical mass from the start; providing little or no ramp up period.

If a large number of companies register across multiple subdomains, the potential benefit of increasing available Domain Names may not materialize. The registration across all available domains would represent an effort to prevent possible piracy and to prevent any potential threat of trademark or tradename dilution. Smaller companies, with more limited resources, may be less able to accomplish these wholesale registrations across subdomains. This inability to register across all domains by smaller companies may create new available names, but may also create some opportunities for piracy.

The USCIB suggests that a cost/benefit analysis should be undertaken to try to quantify the potential benefit that new subdomains could create and compare that to the increased overhead which companies might face in registering, monitoring and enforcing their trade names and marks across new domains. To our knowledge, no such study has been undertaken. While members agree that such a study would be desirable, a few members are concerned that such a study might needlessly delay the evaluation and possible implementation of the IAHC proposal.

The IAHC has set forth the most comprehensive dispute resolution policy and mechanism of any proposal to date, but questions still remain as to whether the WIPO arbitration procedures will provide adequate enforcement and remedies for companies whose trademarks are infringed. The USCIB would request more specifics on the operation and guidelines of the system. The USCIB is concerned that many businesses have not focused on this dispute resolution issue in sufficient detail.

While the IAHC process has been open for comment, it has progressed at a fast pace and has covered an issue which many businesses did not track closely. The Domain Name issue is now gaining more prominence and is being given higher level review by many businesses. The USCIB believes that it is important for as many details as possible concerning this and other proposals to be made available.

5. A Possible Solution

While we have raised potential problems with the IAHC proposal, and understand that other proposals may raise similar or greater concerns, we presume that the de facto monopoly of NSI is not internationally acceptable in the long run. Any alternative solution must recognize the experience and expertise which NSI has gained through their investments of time and effort. The USCIB would urge collaboration between NSI and the IAHC (or whatever proposal gains international acceptance) to provide the best possible solution. NSI's experience would be useful in developing scaleable registration systems and it can provide access to the necessary registration database information to provide for true interoperability. Obviously, a solution of this nature rests on the desire of both parties to work constructively with each other for the benefit of the Internet community. While both NSI and IAHC/CORE seem dedicated to minimizing any chance of fragmentation, many members believe that a collaborative solution provides the best alternative to avoiding the potential for fragmentation which might result if competing Domain Name Systems were deployed. Thus, over time and with proven systems from IAHC/CORE, NSI will cede its de facto monopoly over .com, .org and .net and take its place among the international registrars.

###

Number: 411
From: "Michael Garretson" <eti@teleport.com>
To: NTIADC40.NTIAHQ40(dns)
Date: 8/19/97 4:17pm
Subject: Top Level Domain changes


I support the addition of the seven new top level domains. I also support
the opening of several options for domain registration, instead of just
InterNIC, so long as they are well coordinated and have an international
arbitration methodology.


I would also like to suggest that the registration of multiple "flavors" of
a given domain type be discouraged. This would be one entity registering
"fred.com," "fred.net," "fred.org," "fred.web" and so on through the
top-level domains.


Any search through the phone book is going to find Broadway Books, Broadway
Flowers, Broadway Guitars, etc. -- So much as possible, the domains should
be split between the different companies. While I know some will say
keeping them with one company will reduce misidentification, I hope the
case above will be presented against that, as well.

Thank You,

Michael Garretson
Portland, Oregon, USA

###

Number: 412

From: "David Nash" <5NASHD@stu.jmls.edu>
To: NTIADC40.NTIAHQ40(dns)
Date: 8/19/97 10:58am
Subject: Response to Domain Name Comments

* This message contains the file 'doc_resp.doc', which has been
* uuencoded. If you are using Pegasus Mail, then you can use
* the browser's eXtract function to lift the original contents
* out to a file, otherwise you will have to extract the message
* and uudecode it manually.

Response to Request for Comments on the Registration

and Administration of Internet Domain Names

David B. Nash, III⁽¹⁾

Terminology:

The Request for Comments on the Registration and Administration of Internet Domain Names did not define the term "generic top-level domain" ("gTLD"). In the following responses, I use the definition adopted by the International Ad Hoc Committee.⁽²⁾

Generic top-level domains are referred to as generic because any organization from any country can register a name under them (i.e., .com, .org., and .net).⁽³⁾

In addition, I use "trademark" or "mark" to mean both trademarks and service marks, whether registered or not. Finally, although it may seem a bit jingoistic, I only use English in our examples of possible new TLDs. This is not because I think that English will be or should be the exclusive language of the gTLDs, but rather because it is the language with which I am most familiar.

Introduction:

It should be made clear at the outset: a domain name can be a trademark and a domain name can infringe a trademark. No amount of wishing, posturing, or denying will change that. The United States Patent and Trademark Office will register a domain name like "obesity.com" as a trademark. In Comp Examiner Agency v. Juris, 1996 WL 37660 (C.D. Cal. 1996), the court granted a preliminary injunction because there was a strong probability that the second-level domain name "juris" infringed the registered trademark JURIS. In Hasbro v. Internet Entertainment Group, 40 USPQ2d 1479 (W.D. Wa. 1996), the court also granted a preliminary injunction because there was a strong possibility that the second-level domain name "candyland" diluted the famous mark CANDYLAND. Both of these cases were decided in early 1996. By late 1996, two courts had decided that domain names could infringe trademarks. In both Intermatic v. Toeppen, 947 F. Supp. 1227 (N.D. Ill. 1996), and Panavision v. Toeppen, 945 F. Supp. 1296 (C.D. Cal. 1996), the court held that registering the name of a famous trademark with the intent of selling the domain name to the trademark's owner is infringement of the trademark. Although the precise boundaries of protection for trademarks with respect to Internet domain names is not yet decided, it is clear that some measure of protection exists.

It is unwise to discuss possible solutions to the conflict between the domain name system and trademarks in a vacuum as though the issue has not been pondered before. The IAHC report and its critics, "renegade" TLDs and roots (sometimes referred to as "confederations"), and others have all considered the problem. The IAHC has spent much time and effort in crafting a possible solution to the problem. Although there are many concerns with the IAHC report's approach, it has some useful themes. The first is that the creation of new TLDs should not occur chaotically, but in a controlled manner. The second is the creation of an administrative organization for the Internet to complement the technical ones. Engineers have been able to gather together in the past to keep the Internet running with "rough consensus and running code." However, the future of the Internet depends on a robust administrative organization. The IAHC report has made the important recognition that the administrative and technical needs of the Internet must be met by different organizations. Administrators should set policy. Engineers should implement policy.

A second group of people have set up "renegade" TLDs and roots. The current DNS system depends on about ten root servers that only point to IANA approved top-level domains.⁽⁴⁾

This has not prevented people from attempting to set up their own set of root servers and top-level domains. These "renegade" top-level domains are not accessible to the vast majority of the users on the Internet. This is because the IANA approved root servers are part of the "rough consensus and running code" and the "renegade" servers are not. This is a problem for the future. Whether the solution will be to only allow the government approved roots or to only provide government protection for registrants in government approved top-level domains, is an argument for another day. The important thing to keep in mind is that anyone can set up a group of root servers and TLDs and try to convince the world's ISPs to use them. As long as these "renegade" TLDs and roots receive only a tiny fraction of Internet traffic, they can be safely ignored by trademark owners and governments. That may change in the future.

Other proposals fall into three categories. The first category proposes the creation of a .tm top-level domain or even top-level domains that are specific trademarks. This will not solve the trademark problem, it will just move the trademark conflicts to another level in the DNS hierarchy. The second category proposes completely replacing meaningful domain names with IP numbers or gibberish. This destroys any usefulness the DNS has in organizing the Internet. The third category proposes eliminating gTLDs and having only two-letter national TLDs. Again, this just moves trademark conflicts to another level in the DNS hierarchy. However, it also creates the possibility that an organization with domain names in more than one national TLD will have to conform to many different registration and dispute policies.

An ideal domain name registration system must solve five problems. The first two, concurrent use and dilution, are problems of trademark law. The other three, conglomerates and holding companies, the registration of company names versus trademarks, and generic terms are special cases in name registration.

Concurrent use of trademarks means that more than one organization can use the same mark. This can happen because the marks are used on different types of products, or in different geographic areas, or both. One of the widely used examples is the mark UNITED which is simultaneously used by United Airlines, United Van Lines, and an innumerable number of smaller businesses. All of these business can use this mark, even register it with the federal government, without legal conflict. An ideal domain name registration system will allow all of these businesses the opportunity to incorporate their mark into their domain name.

"Famous" trademarks are protected from trademark dilution under federal law. Trademark dilution is the legal theory that the holder of a famous trademark is injured if that mark is used by any other company in any geographic area. This is because when the famous mark is seen, more than one product or service will come to mind. This means that for certain trademarks (the legislative history of the federal law lists KODAK and BUICK as examples of famous trademarks) only one organization can use the mark. An ideal domain name registration system would allow only the rightful owner of the famous mark and no other to register a domain name incorporating the mark.

The first special case in name registration is the problem of conglomerates and holding companies. These organizations have divisions in different, and often completely unrelated, lines of business. Often, the name of the conglomerate never appears on the products of its divisions and consumers are often never aware of which products or services are ultimately owned by a conglomerate. Therefore, a conglomerate and each of its divisions is likely to be thought of as completely separate entities by the public. An ideal domain name registration system would allow each conglomerate to register a domain name incorporating its company name as well as domain names for each of the companies that it owns.

The second special case in name registration is the problem of the registration of company names versus the company's trademarks. This is very similar to the problem of conglomerates. It is much easier to remember the name of a product or service than to remember the name of the company that provides it. An ideal domain name registration system would allow each company to register a domain name incorporating its company name as well as domain names for each of the marks it uses on its products.

The third special case in name registration is generic terms. Generic terms should be reserved for use only at the gTLD level or to organize domain names. An ideal domain name registration system would prohibit the registration of generic terms as domain names. An example of an ideal use would be .television as a top-level domain. An example of prohibited use would be "television" as a domain name.

The reformation of the domain name registration system should move toward this ideal system. This reformation should not limited to just the top-level of the domain name system. An entire organizational structure should be created. For the example name "chicago-cubs.baseball.professional.teams.sports," the name granting authority should create the structure .baseball.professional.teams.sports, and then organizations that are professional baseball teams would be allowed to register their name under this structure. In addition, tight control would be kept over what name would be registered. Just registering "cubs" under the structure will not be enough because there is a more specific name available "chicago-cubs."

B.1. What are the advantages and disadvantages of current domain name registration systems?

There is one advantage to the current gTLD registration system: with the InterNIC⁽⁵⁾

being the only registrar of domain names under the gTLDs, there is a consistent registration and dispute policy for all of the gTLDs. A person with domain names in more than one gTLD does not have to keep track of different policies and requirements for the different gTLDs.

There are at least three disadvantages to the current gTLD registration system. The first disadvantage is it is not clear where authority resides to create and allocate domain names. Given the current tension between NSF and NSI, a definitive statement about the ownership of the gTLDs must be made by the U.S. government. I recommend that the U.S. government agree with the IAHC report when it states that the top-level domain name space "is a public resource and subject to the public trust." Then the U.S. government, in conjunction with other governments, should create an administrative organization to manage that public resource. The second disadvantage is that there is no competition between registrars. Even though I am recommending that the registrars should not have exclusive control over a gTLD, competition should create better services at better prices. The third disadvantage is the small number of gTLDs and the requirement that each second-level domain name under each gTLD be unique means that, at most, only three organizations can have the same second-level domain name in the gTLDs (e.g., acme.com, acme.org, and acme.net).

B.3. By what entity, entities, or types of entities should current domain name systems be administered? What should the makeup of such an entity be?

B.4. . . . . What is the proper role of national or international governmental/non-governmental organizations, if any, in national and international domain name registration systems?

Policy administration of the domain name registration system needs to be separated from the technical administration of the system. For purposes of this response, the policy administration organization will be called the name granting authority, and the technical administration will reside in the registrars. The name granting authority should be an international, governmental organization with administrative, legal, political, and economic expertise at its disposal. (The registrars will be fully discussed in the answer to D.15.)

The name granting authority would have two major functions. The first would be to create new gTLDs with the purpose of organizing the domain name space in a useful way. (What those new gTLDs should be is further discussed in the response to C.11.). The second is to avoid trademark conflicts. This would be accomplished by evaluating applications for domain names to determine what domain name should be granted, and under which gTLD it should be placed. This is a different approach to registering domain names than under the current system. The idea is to avoid trademark conflicts before they occur, rather than having names registered first-come, first-served followed by administrative hearings and lawsuits.

This approach has two layers to help avoid trademark conflicts. The first layer is to examine the domain name itself. The domain name should not be granted if it is a famous trademark owned by someone else. This prevents possible trademark dilution problems. In addition, the domain name should be as specific as possible. For instance, the Chicago Sun-Times should only get approval for the name "chicago-sun-times" (or something similar like "chicagosuntimes"). Any domain name that is less specific can cause a conflict with another paper whose name also is "Sun-Times." The second layer is to determine where the name can be registered in the domain name organization. In the previous example, the name "chicago-sun-times" should only be allowed to be registered under .newspapers or something similar. To enforce its decisions, the name granting authority would issue a certificate, with the name and location(s) where the name can be registered. The registrars would only be allowed to put a name into the DNS database that had a current certificate.

B.5. Should generic top-level domains (gTLDs), (e.g., .com), be retired from circulation? Should geographic or country codes (e.g., .US) be required? If so, what should happen to the .com registry? Are gTLD management issues separable from questions about International Standards Organization (ISO) country code domains?

The ideal would be to close the current gTLDs to future applicants. However, even if they are not, as .com, .org, and .net get more crowded and crazy and litigious, many organizations will be looking for shelter in the form of a TLD to register under that has a very low probability of resulting in trademark conflict litigation. In addition, people already are used to the idea of looking for a particular organization under a particular top-level domain name--they look for businesses under .com. It is a small step to get used to the idea of top-level domains that differentiate between businesses.

If the current gTLDs are closed to future applicants, they should not be deleted from the domain name system immediately, if at all. Many organizations have spent a great deal of money cultivating awareness of a particular domain name. However, if the current gTLDs are to be deleted from the domain name system, then there should be a period of time to allow a transition to the new gTLDs.

The worst idea in this whole debate is to force the current gTLDs under the .US top-level domain or to eliminate gTLDs entirely and require domain names to only be registered under the two-letter national top-level domains. This, without more, would provide absolutely no relief to the problem of trademark conflicts. It would just change the disputed domain name from xyzcompany.com to xyzcompany.com.us. If both companies in the dispute are U.S. companies or if the trademark is famous, then a conflict still exists. In addition, registration under the current organization of the .US top-level domain would require national companies to choose one city, county, and state to register their name in even though they have operations nationwide.

C.10. Are their technical, practical, and/or policy considerations that constrain the total number of different gTLDs that can be created?

As a general matter, much thought and planning should occur before new gTLDs are created. The ultimate, though unhelpful, answer is that no more gTLDs should be created than are necessary to enhance the usability of the Internet. Only those gTLDs that provide structure for the domain name system should be created. However, there are two types of gTLDs that must be avoided. The first type is the trademark specific gTLD. The second type is any .com clone.

It is a bad idea to allow trademark specific top-level domains for several reasons.⁽⁶⁾

First, just like the misguided proposal to move the current gTLDs under .US, this just moves the trademark disputes to a different level in the domain name hierarchy. Any trademark disputes that would have occurred under .com would occur instead directly under the root. Second, this destroys any type of organizational merit that the domain name system has. The current system has the benefit or organizing domain names into specific categories. Allowing trademark holders to create their own top-level domains creates two classes of businesses on the Internet: those with a vanity TLD and those that are left under .com. Third, it clutters up the top-level name space with names that are more logically and efficiently placed elsewhere.

Another type of gTLDs that should be avoided is hundreds of new gTLDs that are .com clones. One can easily imagine a never ending string of .com, .biz, .firm, .store, .corp, .inc, .shop, .retail, and even something like .businessincorporatedintheunitedstates.⁽⁷⁾

For purposes of U.S. trademark law, however, there is very little distinction between xyzcompany.com and xyzcompany.biz. If each company is in the same line of business, there is a strong likelihood of confusion that leads to infringement of the trademark. Therefore, creating .com clones does not actually increase the number of organizations that can use the same domain name. Even if a new rule is implemented that limits an organization to one domain name (in an attempt to prevent a trademark holder from registering the mark as a domain name in each new .com clone), no one else may get to use the name because of trademark infringement implications.

C.11. Should additional gTLDs be created?

Yes. The creation of new gTLDs should begin immediately. However, they cannot just be created without any organization or forethought. The new gTLDs should create an organizational structure that describes lines of business. For instance, there could be an xyzcompany.recorded.music and an xyzcompany.hardware.computer. These two names are very different and is not likely to cause likelihood of confusion trademark problems.

An argument has been made that line of business categories are not the way to go for several reasons. One reason is that companies change lines of business. Another is that the categories do not accurately reflect modern business. A third is that the categories are so broad that even within each category there might be overlapping marks. The company should have to show on a regular basis that they continue to operate in a particular line of business. The traditional classification systems used for trademarks are too broad to eliminate trademark conflicts. However, any good sized yellow pages has a much narrower classification system. Some of the categories in the yellow pages are so specific that they only contain two or three listings. To prevent the change in business problems, there should be a reexamination part of the renewal process. Of course, some marks will have to be excluded entirely from the gTLDs. Strictly regional marks should be registered under the national TLDs.

Another way to prevent trademark conflicts is to control which top-level domains an individual can register a name in for personal use. Many trademarks are also family names. Keeping the two separate will reduce conflict. The IAHC's proposed .nom takes a step in this direction, although it will probably have to be geographically segmented to be truly useful.

D.15. Should a gTLD registrar have exclusive control over a particular gTLD?

No. There should be shared registries. The company, once it has received a certificate from the name granting authority, should be able to choose between competing registries. Since the company cannot choose where in the hierarchy the domain name will go, having registries with exclusive control over certain top-level domains would force the company to use the registry that had exclusive control over the top-level domain into which they were assigned. Any meaningful level of competition would allow all registries to accept all possible domain names in any gTLD.

E.21. What trademark rights (e.g., registered trademarks, common law trademarks, geographic indications, etc.) if any should be protected on the Internet vis-a-vis domain names?

All trademark rights should be protected. The question is by whom and how to minimize the conflicts. Trademark rights must be protected by the court system of the nation in which the mark is registered or used. The way to minimize conflicts is to have the DNS mimic the way marks are used in the physical world by only allowing marks to be registered in lines of business in which they are used. For example, United Airlines should only be allowed to register something like "united.passenger.airlines." This limits the extent of the mark on the Internet the same way a court or other observer would limit it in the physical world.

E.22. Should some process of preliminary review of an application for registration of a domain name be required, before allocation, to determine it if conflicts with a trademark, trade name, a geographic indication, etc.? If so, what standards should be used? Who should conduct the preliminary review? If a conflict is found, what should be done, e.g., domain name applicant and/or trademark owner notified of the conflict? Automatic referral to dispute settlement?

The only preliminary review should be made concerning into which category the mark should be placed. Once that is determined, the name should be granted on a first-come first-served bases. If there is a conflict, the some mediation process should be available through the name granting authority. However, national courts should always be available as a last resort.

E.23. Aside from a preliminary review process, how should trademark rights be protected on the Internet vis-a-vis domain names? What entity(ies), if any, should resolve disputes? Are national courts the only appropriate forum for such disputes? Specifically, is there a role for national/international governmental/nongovernmental organizations?

National courts are the only legitimate fora for ultimately resolving trademark conflicts. Trademark law is not international law yet, although it has been suggested that the domain name trademark conflicts will spur the development of an international trademark law. Since trademark rights are defined by national law, national courts need to be used to protect those rights.

Preferably an international governmental organization will be the name granting authority. This would solidify the legal authority for the creation and allocation of domain names. The authority should be international in scope to take the Internet's global nature into account. It should be governmental to be legitimate. This authority would create the classifications to determine under which classification a particular applicant fell. An appeals process is needed for the determination of which classifications under which the applicant will be allowed to register a name. If an applicant's mark legitimately can be placed in more than one category, then the applicant should receive a certificate for each category under which the name can be registered.

E.25. Should domain name applicants be required to demonstrate that they have a basis for requesting a particular domain name? If so, what information should be supplied? Who should evaluate the information? On what basis of criteria?

In order for a company to register a domain name, it should have to supply the name granting organization with the requested domain name, why it wants the name (e.g., it is a trademark), and some evidence that would help the name granting authority determine where it belongs in the domain name hierarchy. The name should have to be re-registered regularly. When re-registering, the organization should have to show that the name is still a name or mark used in the line of business that corresponds to the part of the hierarchy it is registered under. There has to be a way to keep track of the organizations registered in each top-level domain. If someone wants to register their personal name in .computer.consulting, they should have to show some evidence that they are in that business and continue to show current evidence every time the name is renewed.

E.27. Where there are valid, but conflicting trademark rights for a single domain name, are their any technological solutions?

Although the technical details will have to be left to others, I have some ideas on what a technological solution as an extension of, or replacement to, the current domain name system, as opposed to an administrative reformation of the current domain name system, should attempt to accomplish. The most important thing is for a person or organization to be found easily in a way that does not cause confusion among users as to exactly which person or organization has been contacted. This could be done through a meta-DNS system.⁽⁸⁾

The meta-DNS system would allow a user to type in "acme" and receive a list of all the organizations that have registered that name under a particular top-level domain and a description of the organization that registered the name. Although the meta-DNS proposal was drafted as a solution to be imposed onto the .com top-level domain, I don't think that is a good idea. It would be overwhelming to have every company registered in the .com top-level domain and then, when the user enters "acme," the user gets a list a mile long of all the companies that use that mark in any manner in any line of business. I think the better solution is to divide the name space into very specific categories based on lines of business, and then use the meta-DNS as a last resort to clean up any lingering trademark conflict problems within the categories. This way, the user receives a much shorter list of the organizations using the mark and all of the organizations on the list will be in the line of business that the user sought after.

E.28. Are there any other issues that should be addressed in this area?

Domain name depth should be emphasized. For example, names should not be registered under .baseball, but under .baseball.professional.teams.sports. This provides a much more complete description of the domain name. This is a characteristic that has been a part of (the civilized portions of) Usenet for years.

1.

^* David B. Nash, III is the author of Orderly Expansion of the International Top-Level Domains: Concurrent Trademark Users Need a Way Out of the Internet Trademark Quagmire, 15 J. Marshall J. Computer & Info. L. 521.

2.

¹ IAHC, Final Report of the International Ad Hoc Committee: Recommendations for Administration and Management of gTLDs (Feb. 4, 1997) <http://www.iahc.org/draft-iahc-recommend-00.html>.

3.

² There are three other types of top-level domains. National top-level domains are only defined by a ISO-3166 two-letter country code (e.g., .us, .au, .de, etc.). International top-level domains are those domains specifically reserved for international organizations (i.e., .int). Finally, special top-level domains are those that have strong, historical restrictions on their use and are used only by organizations in the United States (i.e., .edu, .gov, and .mil). See A. M. Rutkowski, Internet Domain Names and Other Identifiers: A Roadmap Among Issues and Initiatives (Nov. 20, 1996) <http://www.wia.org/pub/identifiers/issues-roadmap.html>. The IAHC report puts these special top-level domains in the national TLD category, presumably because only organizations from the United States can register under them. However, this classification ignores the fact that, unlike the two-letter national domains where any organization in the nation can register under them, only very specific organizations can register under the special TLDs.

4.

³ For an overview of the domain name system, see Neil Randall, How DNS Servers Work, PC Magazine, Sept. 24, 1996, at 217.

5.

⁴ Although there is much confusion and speculation over what will happen when the current contract between the National Science Foundation ("NSF") and Network Solutions, Inc. ("NSI") expires, as of this writing NSI has a contract to perform the domain name registration functions of the InterNIC. Until such time as NSF dissolves the InterNIC and completely leaves domain name registration in the hands of NSI and even though the

InterNIC performs other functions for the Internet, the authors think that it is more accurate to refer to the domain name registrar as the InterNIC.

6.

⁵ One of the "renegade" gTLD registries already claims to have created a .toyota top-level domain.

7.

⁶ "Technically, [a domain name] can have 128 characters with 64 being to the left of the dot and 63 to the right of the dot." David Holtzman, Domain Names: Will We Run Out? (June, 1997) <http://rs.internic.net/nic-support/nicnews/june97/endless.html>.

8.

⁷ Cite this.

###

Number: 413

From: "Paul E. Niedermeyer" <paul@tabnet.net>
To: NTIADC40.NTIAHQ40(dns)
Date: 8/20/97 12:42am
Subject: Request For Comments <Resend>

 Thank you for allowing us to resend our prior copy from yesterday. Please contact me if you have additional questions or problems.

-Paul Niedermeyer

TABNet

Comments on the Registration and Administration of Internet Domain Names

###

Number: 414

From: <chris@creanet.net>
To: NTIADC40.NTIAHQ40(dns)
Date: 8/20/97 3:21pm

PETITION TO THE US DEPARTMENT OF COMMERCE IN SUPPORT OF
THE NAME.SPACE(TM) SYSTEM OF GLOBAL DIRECTORY SERVICES


(The New Paradigm for the Old DNS)


I Christine TREGUIER do hereby support the design of the
expanded toplevel Internet namespace which is currently operated by
pgMedia, Inc.'s NAME.SPACE(TM) service, located on the internet at
http://namespace.pgmedia.net (or http://name.space).


The paradigm implemented by NAME.SPACE(TM) is the most
pro-competitive, democratic and open system proposed so far with
respect to opening up the administration and operation of the
Domain-Name-System ("DNS"). The structure advocated by NAME.SPACE(TM)
removes the artificial barriers to entry that exist today as a result of
the monopolistic control over the domain name registration market
exerted by Network Solutions, Inc. ("NSI"). The NAME.SPACE(TM) paradigm
incorporates a fair, competitive structure which encourages
investment and innovation by companies wishing to compete in the
provision of this service which is essential to the operation and
continued growth of the Internet.


pgMedia, Inc. has created, through substantial private investment in
research and development, its NAME.SPACE(TM) registry administered by
thirteen toplevel root-directory servers located in five countries. The
NAME.SPACE(TM) registry uses innovative and creative techniques which
bring the old DNS out of the Cold War and into The 90'S.


The NAME.SPACE(TM) system decentralizes the administration of DNS and
enables open competition in the Public Domain Toplevel Namespace
without regulation by any governments or quasi-governmental
authority, nor does it require the enactment of new laws or regulations.


Description of the NAME.SPACE(TM) service:


On the NAME.SPACE(TM) system, name registrations are taken by
registrars who administer client accounts under the given toplevel
name categories (publicly shared toplevel namespace). All registrars
must register their digital ID with a trusted third party/parties which
authenticates and authorizes them to function as registries. The
application process is administered by an independent company, similar
to the process used by banks when authorizing merchant credit-card
accounts, and the operation of secure servers used in commercial
transactions on the Internet today.


Registries update the database on demand based on the availability of a
given name address using the IDSD system (IDSD=Integral Database
Synchronizer Daemon), a secure protocol developed by pgMedia which is
available, without limitation or charge. (A detailed description of the
IDSD protocol can be found at http://namespace.xs2.net/IDSD). IDSD
makes it technically feasible for ALL registries to share the toplevel
namespace equally, eliminating any technical justification for
"exclusive" control over any given toplevel name by a single registry,
such as NSI currently enjoys with ".com".


Registration is accomplished instantaneously through an interactive,
form-based interface on the World Wide Web with online payment
options via a secure server. During the registration process, a
registrant establishes an account, a contact "handle" and, of course its
"name". The registrant has the option to choose whether or not its
personal contact information will be publicly listed. All other account
information, of course, remains confidential. The registrant may then
establish a Portable Address Record, over which it has full
administrative access on the NAME.SPACE nameservers. This service
allows a registrant to change service providers and easily take its
"name" to a new host without delay or complications. Upon completion
of the registration process by the registrant, the NAME.SPACE(TM) system
immediately processes the information and creates the second level
entry into the toplevel database, which is then distributed to all other
root-servers via the IDSD protocol. The registration process and the
creation of Portable Address Records are instantaneous, and function
on the Internet within minutes, not days or weeks as in the current
system.


Issues and Answers


Under the NAME.SPACE(TM) paradigm, the toplevel namespace functions as
a Global Directory Service and would be managed within the
competitive marketplace in the general interest of the Internet public
through the various independent registrars. Each generic TLD ("gTLD") is
administered by all registrars who wish to offer services thereunder
with no exclusive claim of ownership of any toplevel name by any
individual, corporation or government, subject to existing intellectual
property law.


These gTLDs may be added or removed based on public demand. Also,
gTLDs may include languages other than English, limited only to the US
ASCII character set, the English alphabet plus 10 digits and the hyphen
for a total of 37 characters.


All leading authorities are in agreement that there is no limit to the
number of possible toplevel names, as there is no limit to the number
of root directories under the UNIX file system. As NSI admits:


"DNS is highly scaleable. There is no technical limit to the number of
new top-level names that could be introduced. The original designer of
DNS, Paul Mockapetris, has verified the scalability of DNS."


(http://rs.internic.net/nic-support/nicnews/jun97/MYTHS4.html)


Thus, any claim that expanding the toplevel namespace is technically
not feasible is simply unfounded. The proponents of such claims seem
to be guided by a desire to limit the potential market so as to create an
artificial scarcity which translate into higher prices and profits.


The use of arbitrarily defined and limited categories such as ".com" has
forced many registrants to engage in verbal gymnastics, and to rely on
unwieldy content-based search engines - this would be obviated by the
full implementation of the NAME.SPACE(TM) paradigm. Thus, for example,
Acme.computers and Acme.plumbing could both have a presence on the
Internet without having to artificially pervert their names. The
"byte-counter mentality," which has plagued us with the dreaded
"Millennium Bug," was responsible for the initial constraints on the
toplevel domain name nomenclature. The NAME.SPACE(TM) system simply
recognizes that such limitations have long since been eliminated and
are wholly artificial.


With respect to intellectual property issues, no regulatory framework
can assure the complete protection of holders of such rights against
infringement by unauthorized parties. However, the potential for such
infringement, which exists in all published media, should not be used as
a basis to limit the free speech rights of the vast majority of law
abiding users of the Internet, while protecting artificial monopolies.
Furthermore, it is wholly inappropriate to empower any registrar to
adjudicate the rights of holders of intellectual property, for that role
must ultimatly reside with the courts.


Fees for registration services should be dictated by the market. Waiver
of fees and discounts should be considered for qualifying educational
and non-profit organizations, as well as a selection of totally free
categories (such as the Free.Zone provided currently by NAME.SPACE(TM)).


In conclusion, NAME.SPACE(TM) has developed and implemented a new
paradigm for the Global Directory Services on the Internet by bringing
the function of the old DNS, a legacy of the Cold War, into sync with the
current dynamic of the public, global, civilian and commercial Internet.


The NAME.SPACE(TM) system is a reality today. The NAME.SPACE(TM)
automated registry has been fully functional for nearly one year now
and has proven its reliability and desirability as evidenced by the
thousands of users who have been using the NAME.SPACE(TM) servers to
resolve their DNS and those who have registered their names in
NAME.SPACE(TM) .


I fully endorse and support the endeavors of pgMedia, Inc. and the
NAME.SPACE(TM) system and highly recommend that the U.S. Department of
Commerce recommend and concur in its full implementation on the
Internet.


Paris 08-20-97

###

Number: 415

From: "Timothy Sowers" <TLSvideo@classic.msn.com>
To: NTIADC40.NTIAHQ40(dns)
Date: 8/17/97 11:08pm
Subject: Domain Names

Greetings:

In regards to the Domain Name Conflict; If the Internet is to grow in an
organized manner then new domain names must eventually be added. However, the
fact that a company or organization can hold a monopoly on domain names, thus
slowing Internet growth, is idiotic.

Sincerily,
T.L. Sowers

###

Number: 416

From: Claude Potvin <cpotvin@scc.ca>
To: "'webmaster@ntia.doc.gov'" <webmaster@ntia.doc.gov...
Date: 8/22/97 10:37am
Subject: Domain names

Good morning.

Please forgive me if you are not the correct individual that should be receiving these comments. I would be very grateful if you could forward this message to the correct person.

Working within an information service department, I've seen an increase in E-mail and Internet inquiries. In addition to servicing our outside customers we also answer to other that provide funding that allows us to continue providing these services. For that reason, we maintain statistical information on who has contacted us, where the inquiries originated from (geographically) and what the inquired intailed. We depend allot on the mailing address or the area code of the telephone number for this information.

Unfortunately, with the increased E-mail and Internet communiqués it has made this type of statistical maintenance very difficult.

For this reason I would like to suggest that what ever decision is made that the following is also considered. When assigning a new domain name, the ending is designated to assist in identifying where the caller originated from i.e. xxxxxxxx@xxxxx.xxxx.xxx.ON.CA where as the two character representation of provinces or states and country could be used. Designation such as COM, ORG, TEL (for telecommunication industries), EDU, GOV, PROV (for provincial governments), etc..... could be assigned just before the provinces, states and country codes.

I would like to thank you in advance for allowing me this opportunity to express my opinions.

Sincerely,

Claude Potvin

###

Number: 417

From: <Michael.Hardy@ATONAT.ausgovtax.telememo.au>
To: NTIADC40.NTIAHQ40(dns)
Date: 8/26/97 2:25am
Subject: RFC Domain Names

Dear Ms Washington,

Attached is a response from the OECD cybertax group in
relation to your RFC on Domain Names.

I apologise for this late response. However, some of they key
members of the cybertax group have not become available until
the last week and I have been having some difficulty
contacting you by e-mail.

The attached document is in a Word for Windows 2 format.

If you have any queries, or difficulties, please do not
hesitate to contact me.

Michael Hardy.

Michael Hardy
Chair
OECD Cytbertax Group
C/- Australian Taxation Office
2 Constitution Ave
Canberra ACT 2600
AUSTRALIA

25 August 1997

Ms Partrice Washington
Office of Public Affairs
National Telecommunications and Information Administration (NTIA),
Room 4898
14th St and Constitution Ave, NW
Washington DC 20230

Dear Ms Washington,

Request for Comments on the Registration and Administration of Domain Names

I work for the Australian Taxation Office and I am the chair of the Organisation for Economic and Cultural Development (OECD) "Cybertax" group, which is examining the implications of Internet commerce for tax administration, particularly tax evasion and tax avoidance.

I am writing to you in my OECD capacity with some comments on the registration and administration of domain names.

Generally we believe that domain name registration should be subject to appropriate, but not over-burdensome, regulation to ensure consumer and commercial protection and the integrity of national laws. In may respects domain name registration might be seen as analogous to the registration of business names in the physical commerce environment and subject to broadly comparable requirements.

Areas of general concern with the current arrangements are that their is an apparent weakness in the relationship between the country codes and the parties actually operating an Internet site and a lack of clarity as to which enterprises are appropriate under a given gTLD.

Question 1, Section B General/Organisational Framework Issues

The current domain name registration system is useful in providing both an indicator of the type of Internet presence (eg. .COM, .GOV etc) and the nominal geographic nexus (eg. .AU, .FR, .NL etc).

Some of the disadvantages, from a tax administrator's perspective, are that there is not always a strong relationship between reality and what is intimated by a domain name.

Question 5, Section B General/Organisational Framework Issues

Because of the overall usefulness of the gTLDs and country codes, we would not like to see either element retired. Should the current domain name registration system be revised which necessitates the retirement of gTLDs or country codes, we would be interested in discussing mechanisms by which this information might be available under any alternative system.

Question 11, Section C Creation of new gTLDs

While it would be technically possible to create new gTLDs, perhaps there should be an examination of the existing gTLD arrangements, particularly in the area of providing a guidance framework as to which Internet activities are appropriate under each of the existing gTLDs and some attempt to restructure revise existing enterprise and domain name pairings within such a framework.

Question 15, Section D, Policies for Registries

If the question is intended to imply that a particular domain name will not be issues on an exclusive use basis and might be used by two or more enterprises simultaneously then, as tax administrators, we would have grave concerns over the impact this would have on commercial identity on the Internet.

If however, the question is about assigning a registrar exclusive rights to issue a particular series of domain names, then we have no comment.

Question 20, Section D, Policies for Registries

Obviously domain name registrars should comply with relevant national laws and, in particular, should provide registration information to tax administrations as required by law.

If you have any queries, please do not hesitate to contact me, in writing, by facsimile on +61 2 62161591 or email at Michael.Hardy@atonat.ausgovtax.telememo.au

###

Number: 418

VIA COURIER

###

Number: 419

From: scot mcphee <smcp@hotmail.com>
To: NTIADC40.NTIAHQ40(dns)
Date: 8/24/00 11:58pm
Subject: Creation of New gTLDs

I have a simple suggestion for the creation of gTLDs;

For a start, the system is too US centric. The US Commerce Department
should concern itself with the .us domain only. Only international
bodies should deal with "generic" TLDs. Has it occurred to anyone that
the USA has no sole rights on the Internet anymore?

As for what to DO;

Purge .com, .org, .net etc of all trademark registrations. (IMPORTANT
and very UNPOPULAR but NECESSARY step!).

Create gTLDs for each of the International Trademark registration
categories. Trademark registrations in each of these domains would be
allowed only on production of the requisite TM registration papers and
the mark would go into the category it is registered for. Then these
registrations are also sorted into something like product categories.

.com registration would be allowed only for genuine COMPANIES,
appropriate documentation would be required to be furnished for
registration in .com to proceed, the registration name would have to
conincide some way with the actual company name.

This is the way .com.au is run. I think it works well. It also clears
"com" out of all the bullshit domains that get registered.

Perhaps, then, the .com reg would come automatically on company
registration and the TM registrations likewise.

You should also create ONE TLD with a "free-for-all" type philosophy so
people can register their stupid name registrations at will.

Also, instead of concentrating on the best interests of REGISTRANTS why
not concentrate on the best interests of USERS trying to navigate this
mess.

Lastly, DON'T RUSH CHANGES. The existing system works OK. If you rush
it, you fuck it for EVERYONE not just the US or a few companies.

###

Number: 420

COMMENTS OF DUN & BRADSTREET ON THE