From: Shari Steele <>
To: NTIADC40.NTIAHQ40(dns)
Date: 3/23/98 5:42pm
Subject: EFF's Comments on Domain Names

Shari Steele, Staff Attorney
Electronic Frontier Foundation
1550 Bryant Street, Suite 725
San Francisco, CA 94103

March 23, 1998

Karen Rose
Office of International Affairs
National Telecommunications and Information Administration (NTIA)
Room 4701
U.S. Department of Commerce
14th Street and Constitution Avenue, NW
Washington, DC 20230

Sent via e-mail ( and Federal Express.

Dear Ms. Rose,

The Electronic Frontier Foundation (EFF) thanks you for this opportunity
to comment on the Department of Commerce (DoC)*s attempts to find
consensus on Internet domain name policy. EFF is a nonprofit,
public-interest organization working to protect rights and promote
responsibility in the electronic world. We have read with interest your
proposal for "Improvement of Technical Management of Internet Names and
Addresses," and our comments are below.

Protection of Human Rights and Free Expression Should Be Overriding

EFF believes in the basic principle that any foundation for governance
of a communications system, such as the Internet, should stand on the
fundamental human right of free expression. The strongest guarantees of
free expression, free association, due process, and nondiscriminatory
administration should be written into the charter of any organization
empowered with Internet oversight. The Internet should be administered
on principles compatible with, but stronger than, the United Nations
Universal Declaration of Human Rights. These guarantees regarding
Internet governance should not permit any party to limit human rights on
the Internet. This guards against a repetition of the practice of some
national governments, which undermine the UN Declaration under the
pretext of "morality, public order and the general welfare" (Article 29
and 30). They should only be amendable by an extraordinary
supermajority procedure, similar to a U.S. Constitutional amendment.

The IANA Is the Best Choice for Overseeing the DNS System

The DoC proposal states as a goal that the U.S. government should end
its role in the Internet domain naming system (DNS) as soon as possible.
EFF could not agree more. We believe that the Internet Assigned Numbers
Authority (IANA), which currently administers the DNS and has been doing
this well for the past 20 years, is the proper organization to continue
administration of the DNS. IANA is a private organization whose only
connection to any government or collection of governments is its history
of U.S. government research funding. The authority of any new structure
designed to continue IANA's job of administrative maintenance of the
Internet should be clearly defined as independent of any and all
governments. The U.S. government should not attempt to exercise
oversight or a veto over a transition of the IANA to a new corporation;
such oversight will be carried out by Internet users. The Internet
community is much more likely to trust a new corporation to manage
Internet administration if it is clearly derived from the IANA and
continues the same personnel and general policies.

Any organization selected to administer the DNS must remain private,
whether in the legal form of a corporation or otherwise, and should be
not-for-profit, operating in the public interest for the benefit of the
Internet. This administering organization should be governed by
democratic procedures which give suffrage to all stakeholders on the
Internet. To the extent that any individuals wield power within the
organization, they must be charged to wield that power for the long-term
benefit of the public, the Internet users, and the entire human race,
especially since all three of these are synonymous in the long term.

It is of critical importance that any administering organization not
simply apply U.S. law, and that any new Internet domain naming system
not rely on U.S. jurisdiction over trademark issues and dispute
resolution processes. The Internet is international in scope, and its
administering body must be truly international in scope.

Network Solutions Cannot Be Trusted With Any Public Resource

EFF is very concerned that Network Solutions is attempting to convert
its five-year contract into a permanent monopoly. EFF believes that the
National Science Foundation (NSF) made a mistake by failing to control
this for-profit company to protect the public interest. Though the
fundamental mistake was made by NSF, it was compounded many times by the
arrogance of Network Solutions' management. Yet, the DoC proposal gives
Network Solutions continued control of the .com, .net and .org domains.

EFF believes that the current management of Network Solutions has shown
a profound disregard for the public interest, which should disqualify it
and its parent company, Science Applications International Corporation
(SAIC), from having any benefit or privilege extended to them in the
future management of the domain name system. They made a deliberate,
intentional and ongoing attempt to steal from the public the resource
they had a five-year stewardship contract to manage and protect. It is
as if they got a contract to repair the stonework at Mount Rushmore and
ended up trying to own the national park.

As a result, they should not be trusted with even temporary control of
any valuable public resource. If Network Solutions or SAIC continue in
the domain name business, they must operate only in highly competitive
parts of the business. In particular, neither Network Solutions, SAIC,
nor any "nonprofit organization" started by them, or with overlapping
directorship, influence, or control by them, should manage the root or
any global top level domain (gTLD). Any changes to the DNS must ensure
that people will continue to be able to register in the .com, .net,
.org and .edu top level domains without any involvement of Network

We were happy to see that the DoC proposal requires Network Solutions to
give the U.S. government copies and documentation of all data, software
and licenses to other intellectual property generated under the
Cooperative Agreement and to turn over control and management of the
main root server to the U.S. government. We are concerned, however,
that the current wording leaves significant loopholes that should be
patched up.

Network Solutions should be required to turn over all root servers that
it operates, not simply the "A" root server, and any separate servers it
uses for the .com, .org, .net and .edu domains. Due to the difficulty
of changing the set of globally known root server addresses, Network
Solutions must relinquish the entire set of IP addresses delegated to
the InterNIC, specifically, which contains both the "A"
and "J" root servers. This should happen not "when the government
directs," as indicated in the DoC proposal, but rather when IANA or the
new corporation directs. Network Solutions should begin immediately
preparing for this by moving other hosts off that network number, and
doing anything else necessary to ready itself for the transfer.

Furthermore, the requirement that Network Solutions turn over
"appropriate licenses to other intellectual property" is far too vague.
All intellectual property generated under the Cooperative Agreement is
work-for-hire, and the government owns it in trust for the public.
Network Solutions has no ownership rights to any of the work it created
under that stewardship contract, so how could it issue licenses? The
government should explicitly put ownership of all such results into the
public domain, so they can be used by the public that paid for them. It
should then make the results accessible via an NSF web site and via
Freedom of Information Act requests. Please ensure that Network
Solutions retains no credible claim to any of the public's intellectual

No For-Profit Corporations or Governments Should Control the Root or

Domain name registration and the generic top level domains themselves
must not be monopolized by a single for-profit registry, treated as any
for-profit entity's intellectual property, or controlled by or from
within any single governmental jurisdiction. These domains have become
international and should not remain a U.S. monopoly. They are a global
public trust and should not be exploited by profit-seeking companies or
for regional advantage.

IANA Should Have Supervisory Control of All gTLD Registration Databases

IANA, acting on behalf of the public, should have supervisory control of
the databases containing the registration data of each global top level
domain in the DNS, including the .com, .net, .org and .edu domains.
IANA should determine when and whether to create new gTLDs, including
specification of the alphanumeric strings, the timing of introduction of
new gTLDs and the number of gTLDS. IANA should ensure that the data in
each database is freely available to the public, subject only to
international legal restrictions relating to privacy.

Each database should operate in the public interest on a cost recovery
basis and not for profit, under the overall supervision of IANA. Access
for registration of second level domains in all databases should be
equally available to all registrars (except in exceptional circumstances
determined by IANA) on a non-discriminatory basis. The administration
of the domain name system and generic top level domain registries must
provide for domain name portability rather than making it difficult or
impossible for a customer to switch registries.

If a de-centralized technical method, which would allow several
different registries to accept names in the same gTLD, can be made to
work, it should be adopted. This would provide a very strong level of
domain name portability, in which a user could change not only the
registrar (retailer) but also the registry (wholesaler) that controls
the pricing and policies for the user's established domain name. This
would provide a much more competitive market, avoiding lock-in at both
levels, and would also permit registries to be operated for profit,
since registries that misserved their customers would be unable to
prevent users from switching to another registry for the same TLD. A
prototype implementation has been suggested to IANA; it involves having
the name servers for a given gTLD independently retrieve the pieces of
the database from the various registries. Each server would merge the
database pieces using freely available software, which would resolve
conflicting registrations since the last merge, before publishing the
new zone database. The prototype would work for small numbers of
registries sharing a zone -- perhaps a dozen -- and could be extended to
larger numbers after gaining experience and evolving the tools. This
proposed method would require oversight by IANA (under threat of removal
from the gTLD delegation) to prevent disruptive or negligent practices
by the registries.

Internet Technical Standards and Administrative Processes Should Not Be
Impeded by Lack of Consensus on Separate Issues Such as International

The current policy of providing ownership rights in Internet domain
names based on trademark registration is flawed and should be abandoned.
No one group of intellectual property holders' interests should outweigh
any other group's, or outweigh any other rights held by individuals and
the public. The current policy with respect to trademarks and domain
names does not even correctly reflect U.S. trademark law, let alone the
laws of other countries or any emerging international law of trademarks.
The policy appears to have been adopted to benefit Network Solutions
(reducing the number of lawsuits against it by large corporations)
rather than to define a balanced public policy. Yet the current
proposal seems to perpetuate this horrible policy.

Under U.S. law, the First Amendment to the Constitution carries
considerably more weight than the Lanham Act. Under the current DNS
administration, trademark holders are given greater rights to domain
names than others with legally recognized interests in specific domains.
This is not a reasonable application of the law, and it is not sound
public policy.

Even the United States Patents and Trademarks Office (PTO) recognizes
that more than one entity can hold a trademark on the same word, name,
or symbol provided that it is used in completely different business
categories. Failure to provide for multiple uses of the same
trademarked character string in a domain name has resulted in big
companies beating up on small companies and individuals using the same
or similar names.

Furthermore, the PTO recognizes that registration of a trademark creates
a right in the mark while the trademark is still pending. The Internet
policy does not recognize registered but still pending trademarks, and
small companies with pending applications have been bullied into giving
up their domain names by larger companies with similar trademarks.

But most importantly, there are other legitimate uses of words, names
and symbols that have nothing to do with trademarks that get and deserve
legal protection yet are not recognized by the current policy. For
example, Kayvan Sylvan, a man who runs his own computer consulting
business, registered the domain name Sylvan Learning
Systems, which holds a trademark on Mr. Sylvan*s last name, threatened
to take that domain name under the current policy. Mr. Sylvan had to
trademark his last name in a foreign country in order to keep his domain
name! There are other examples of nontrademarked, yet legal, uses of
terms. For example, the World Boxing Association might want to use the
domain name, even though Hasbro toys has trademarked the
term knockout as the name of a game. This should be a legally protected
use of a domain name.

Domain names cannot be equated to trademarks or brand names, since the
sole domain name cannot be used by multiple participants to serve their
various non-infringing functions. Instead, domain names should be
distributed on a first-come, first-served basis, and only if the company
that wants a domain name can prove that the current owner is confusing
the customer (i.e., McDonald's Hardware decides to get into the
fast-food business), could it then prove trademark infringement. And
even then, it would not necessarily succeed in taking over the domain,
just forcing the infringer to stop infringing on their trademark.

Trademark concerns cannot burden the free expression uses of domain
names. Users of domain names cannot be required to identify themselves
or provide an address for service of process or any other reason
suggested in Appendix 2 of the DoC proposal in order to get a name, just
as the U.S. government cannot compel speakers in the physical world to
identify themselves. No period of suspension of a domain name
registration, which serves the same function as censorship in this
context, shall be imposed prior to an order by a court of competent
jurisdiction, with the burden of proof being on the censor.

Many of these trademark disputes can be avoided with the creation of a
multitude of additional top level domains. The current policy is flawed
and must be repaired.

Domain Name Ownership Disputes Should Be Resolved by Contract Law
A uniform procedure should be established for resolving domain name
ownership disputes by contract among registrars of second level domains
in each of the gTLDs. This procedure must strike a proper balance,
enforcing the law as interpreted by courts, between domain name holders
and the owners of trademark rights. It should offer an efficient and
inexpensive means of dispute resolution without supplanting or
interfering with the jurisdiction of national courts or the rights of
Internet users to have resort to the courts. Finally, domain name
registries should be insulated from trademark claims. The process of
domain name registration cannot become a proxy for trademark litigation.

Maintenance of the .edu Domain Should Not Be Given to Any Organization
With Exclusionary Intent

The current administration of the .edu domain is exclusionary and geared
to support American four-year universities at the expense of other
educational institutions, including local community colleges. This gTLD
should be run by the same international, non-profit organization
maintaining the other top level domains and should be open to all
educational institutions, as defined by the institutions themselves. A
recent proposal being circulated by Educom is completely unacceptable,
where a hand-picked panel of college presidents and chief information
officers from four-year universities that are members of Educom will
decide which institutions can get a .edu address. Faculty members,
students and non-four year academic institutions should all be included
in any decisions related to the .edu domain.

Payments to the Intellectual Infrastructure Fund Should Be Returned to

EFF was pleased to see that the DoC proposal terminates the Internet
Intellectual Infrastructure Fund (IIIF) as of April 1, 1998. However,
this does not remedy the current problem of the illegal overcharges that
have already been made. The Intellectual Infrastructure Fund should be
fully returned directly to the users who paid in. Such refunds should
be returned to the users, so they can spend them on any domain name
registrar or on other things, rather than being applied as credits for
future domain name service from Network Solutions. Network Solutions
(which instigated the policy of charging users for the IIIF) -- not the
public or domain name users -- should bear the costs for defending
against any lawsuits and administering the refunds.

No Transition Period Is Needed; Current IANA Can Make Decisions

The government's proposed transition plan is flawed in numerous ways.
It should not be adopted. Instead, the current IANA should determine
the short-term evolution of domain name administration, as it has
determined the long-term evolution. As the new IANA corporation is set
up, this function can be transferred to it by the old IANA.

The number of new domains should not be fixed by the U.S. government.
The U.S. government should not define rules for qualifications of a
registrar or a registry, or even define that the two must be separated.
The role of the government should be to drive the process to consensus,
not to dictate which consensus should be arrived at.

Furthermore, many of the requirements for registrars and registries are
poorly thought-out. First, encryption, as opposed to authentication,
should not be required for registrar/registry transactions. Such a
requirement would put U.S. organizations at a disadvantage, since they
cannot export their interface software (which is required by another
part of the proposal, in Appendix 1.1.g).

In addition, registrars do not need multiple connections to the
Internet. This requirement simply raises the base cost of providing
service. Competition will let users pick a registrar that is responsive
and available. Most of the specified criteria for registrars do nothing
more than reduce competition by raising barriers to entry.

Thank you again for giving us the opportunity to comment as you work to
formulate policy in this important area. We would be happy to meet with
you and work to create a system that is in the public interest. Please
contact me at 301/375-8856 if I can be of any further assistance.


Shari Steele
Staff Attorney
Shari Steele, Staff Attorney
Electronic Frontier Foundation 301.375.8856 (v)
P.O. Box 649 301.283.5337 (f)
Bryans Road, MD 20616

CC: NTIADC40.SMTP40("","eff-staff@eff...


From: Glenn B. Manishin <>
To: NTIADC40.NTIAHQ40(dns)
Date: 3/23/98 5:50pm
Subject: Comments of CPSR

Dear Sir/Madam:

On behalf of the Computer Professionals for Social Responsibility,
attached in Microsoft Word 6.0 and ASCII format are electronic files
containing CPSR's comments on the Discussion Draft on Technical
Management of Internet Domain Names (the "Green Paper").

Hard copies of these documents will be delivered by messenger to
Karen Rose of NTIA on Tuesday, March 24.

Please do not hesitate to contact the undersigned counsel for CPSR
if you have any questions or if there is any technical problem with the
binary attachments to this message.

Glenn Manishin

Glenn B. Manishin
Blumenfeld & Cohen-Technology Law Group
1615 M Street, N.W. Suite 700
Washington, DC 20036
202.955.6300 X229 Tel/vms
202.955.6460 Fax

Before the




Washington, D.C. 20230

In the Matter of )
Improvement of Technical Management of ) Docket No. 980212036-8036-01
Internet Names and Addresses; )
Proposed Rule )



The Computer Professionals for Social Responsibility (CPSR) <>, by their attorneys, respectfully submit these comments on the "Green Paper" <> released by the National Telecommunications and Information Administration (NTIA) <> regarding present and future systems for registration and administration of Internet domain names.


NTIA has developed a serious and generally sensible review of Domain Name System (DNS) administration that embraces the key principles establishing a fully non-governmental, self-governance model for the Internet. CPSR is pleased that the U.S. government has recognized the need to act -- as CPSR proposed -- as a catalyst in assisting the creation of the new self-governance organizations that will be necessary to complete the transition to a non-governmentally administered Internet and that can continue the collaborative, consensus-driven processes that have traditionally characterized the development of policy and standards by the Internet community

We strongly believe that the Green Paper's unequivocal endorsement of registrar competition, via shared generic Top Level Domains (gTLDs), is appropriate and in the best interests of all Internet users, developers and service providers. As detailed in our August 1997 comments in this proceeding, CPSR -- a public interest alliance of information technology professionals and others concerned about the impact of computer technology on society, founded in 1981, with over 1,400 members and 22 chapters nationwide -- shares the Green Paper's concern with avoiding changes in DNS administration that could threaten the technical efficiency and stability of the Internet. As a result, we concur with the Green Paper's recommendation that the DNS "root" servers must be coordinated, and that "competing root systems" would represent a tragic departure from the principle of universal accessibility that is inherent in the Internet. As CPSR commented previously:

[C]hanges to the current DNS model must reflect the twin goals of maintaining Internet self-governance, thus minimizing government's substantive role in Internet administration, while avoiding the continuation of de facto DNS monopolies in the increasingly commercialized Internet. . . . The Internet domain name registration process should be opened to competition for all existing and newly created generic top-level domains (gTLDs). . . . The Internet's "root" server administration responsibilities should be coordinated and centralized in order to assure reliability and scalability of the Internet.

Consequently, the Green Paper's proposal for the creation of a "private, not-for-profit corporation . . . to manage the coordinated functions in a stable and open institutional framework" represents a model for Internet administration that is orders of magnitude better than the flawed Council of Registries (CORE) initiative supported by some International quasi-governmental organizations (ITU, WIPO, OECD, etc.) because such organizations should have no formal role in Internet governance or domain name registration. We only caution that in allocating Board of Directors seats on the Ònew corporation,Ó special care must be taken to ensure balanced and fully international representation by all Internet stakeholders, including the substantial segment of non-commercial Internet end users who are not reflected in the Internet Society or other current Internet-related governance organizations. CPSR agrees wholeheartedly with the Green Paper's principle that the process of DNS reform "should, as far as possible, reflect the bottom-up governance that has characterized development of the Internet to date."

At the same time, CPSR has a few significant reservations regarding the Green Paper's specific proposals, especially its attempt at resolution of what the Green Paper terms "the trademark dilemma." We continue to believe that domain registration should be separated from trademark issues. Registrars should not be involved in trademark dispute resolution, but rather should refer all trademark issues to appropriate national and international judicial bodies. Although in comparison to the CORE initiative, the Green Paper's trademark procedures are relatively benign (e.g., domain owner information, searchable database of second-level domain registrants), the Green Paper nonetheless would mandate an "alternative dispute resolution" process" for all registrars. Along with a questionable assertion of unilateral US jurisdiction to dictate -- if only in the short run -- the pace of competition in gTLD creation and registration, which CPSR also opposes, the Green Paper's foray into the quagmire of Internet trademark issues is its greatest weakness. In the absence of any technical reasons to decide trademark issues or to set gTLD constraints, of which CPSR is unaware, the US government should not, of its own authority, attempt to impose such rules on the Internet prior to a consensus being achieved in the new Internet self-governance corporation as to these important matters.

A. Trademarks

Although the Green Paper agrees that creation of a "monolithic trademark dispute resolution mechanism" is inappropriate, its proposed mechanisms for "balancing" rights of domain holders and trademark owners represent a clearly inappropriate intrusion of trademark concerns into DNS administration. Much of the impetus for the CORE plan stemmed from differences between trademark owners and others as to the relative rights, and economic leverage, that NSI's own (and unilaterally imposed) dispute resolution policies have engendered over the past several years. The Green Paper appropriately eschews the highly bureaucratic, centralized procedures and rules developed by CORE to protect the commercial interests of trademark owners.

CPSR does not object to the provisions suggesting that all gTLD registrars should maintain current, accurate records of each second-level domain registrant, and that each registrant should certify that it is not aware of an entity "with superior rights in the domain name." These are reasonable proposals for information collection that will assist in the resolution of the tiny minority of domain name registrations that are ultimately disputed. Yet the Green Paper also proposes to require each registrar to make available -- and agree to abide by -- "a readily available and convenient dispute resolution process." This mandatory alternative dispute resolution (ADR) provision should be deleted. Trademark owners, as commercial entities, have the legal and financial responsibility to "police" use of their marks and, if necessary, take enforcement action under specific national laws in the applicable country of jurisdiction. There is nothing about the Internet that warrants the creation of a dispute resolution mechanism designed, as the Green Paper apparently aspires, to lessen the burden on trademark owners of protecting their intellectual property. If the objective of the Green Paper is, on the other hand, solely to "provide trademark holders with the same rights they have in the physical world," then these ADR provisions are plainly superfluous, as they provide intellectual property rights, and "convenience," to trademark owners that are not enjoyed with respect to any other medium of communication.

In its August 1997 comments, CPSR recommended that gTLD registrars should not intervene in disputes between trademark owners and domain holders, but rather should refer disputes to the applicable national courts. We urged the US government to "assist this sensible result by making clear its view that because registration of a domain name is not the 'use' of a trademark, the domain registration process cannot constitute 'contributory infringement' under US trademark law such that registries would be exposed to legal liability for ignoring a request to 'take down' a potentially infringing registration." Consistent with a subsequent decision by the United States Court of Appeals for the Ninth Circuit, the Green Paper states that "the law provides no basis for holding that a registrar's mere registration of a domain name . . . should expose it to liability." We applaud NTIA for this forthright statement, and strongly suggest that there should be no additional procedures, rules or requirements for domain name registrations involving potential conflicts with commercial trademarks.

The Green Paper asks for comment on two specific proposals regarding trademarks. First, whether registrars should be compelled to provide for an accelerated dispute resolution timeframe, with suspension of a disputed domain name registration in the interim. Second, whether at the time of domain name registration, registrants should be required to agree to judicial jurisdiction "where the registry is domiciled, where the registry database is maintained, or where the "A" root server is maintained." Consistent with the foregoing, CPSR believes that a mandatory suspension and ADR deadline is inappropriate and unnecessary. Moreover, while we agree that clearer rules of Internet-based jurisdiction are needed to deal with the new medium of cyberspace, whether registries should impose jurisdictional conditions on gTLD domain registrations is an issue best resolved, if at all, by means of an international, multilateral treaty-level legal agreement, rather than the unilateral mandate of the US government.

B. Internet Self-Governance

As noted, the Green Paper's proposal for creation of a private, not-for-profit corporation (the new corporation) to manage coordinated Internet functionalities, including the present IANA function of IP number allocation, is a landmark development. If implemented correctly, this approach will help assure that the ongoing DNS administration will be handled in an open, balanced and non-governmental manner, with full participation by consumers and small commercial entities, in addition to trademark owners.

It is essential, however, that the new corporation is truly open and free from conflicts of interest. Accordingly, while CPSR agrees with reservation of Board seats to regional IP number registries (ARIN, APNIC, RIPE) and the IAB, we do not believe that the reservation of only one Board seat for "an individual or entity engaged in non-commercial, not-for-profit use of the Internet" is appropriate. Internet "users" have, unfortunately, for too long been synonymous with entities operating commercial Internet activities, rather than the millions of individuals ("end users" in other parlance) who actually use the Internet for communications. CPSR urges that both the Internet "membership association" contemplated by the Green Paper, as well as the new corporation Board seats, have at least equal representation by both Internet end users and commercial Internet providers. In other words, four of the seven Board seats allocated to a Òmembership association (to be created) representing Internet usersÓ should be reserved for non-commercial, non-profit Internet users. Moreover, Board members should not be employees of registrars, registries or of any other entity competing for commercial services in the domain name registration market.

CPSR concurs that the new corporation must "reflect changes in the constituency of Internet stakeholders." That means, we believe, that governance of the new corporation must be formally vested in an international Board, with full representation from all parts of the world. Both in order to avoid the appearance of US hegemony, as well as to secure the political support the Green Paper requires in the context of the global Internet, CPSR believes that Board representation should also be reserved for international representatives. While these should not be government or quasi-government representatives, the importance of internationally balanced representation cannot be understated.

The new corporation should be prepared to deal with disputes between registrars and registries, but it should not be designing protocols or participating in technical work now done by the IETF. It should also not use its number-assigning role to "pick winners" in opening of new gTLDs. In this regard, while CPSR agrees that in performing coordinated Internet functions the new corporation will be acting "much like" a standards-setting body (and with applicable legal liability if competitive safeguards and open processes are not observed), it is vital to state clearly that the new corporation will not be engaged in developing or adopting Internet technical standards. For instance, if technical standards are required for the development of registry sharing or for root server coordination, these should continue to be driven by the technical members of IETF, rather than the new corporation.

C. Registrars and Registries

CPSR's basic principle, as outlined in its August 1997 comments, is that DNS competition should be introduced in a way that does not jeopardize the technical stability of the Internet. Accordingly, we opposed the CORE proposal because it did not directly deal with the importance of coordination and/or centralization of the root server functions that are essential to maintaining universal resolvability of all gTLDs on the Internet. CPSR also opposed "branded" (i.e., proprietary) gTLDs, and believes that marketplace competition in the registration market will be more than adequate -- as current trends already demonstrate -- to produce innovation, efficiency and capital investment.

The Green Paper's response is for the most part entirely consistent with these views. First, the Green Paper advocates centralization of root server administration in the new corporation, but recognizes that if a "system of authoritative roots" is coordinated and synchronized, there is no technical need to retain a single "dot" server, and with it a continued risk of catastrophic network failures. Second, the Green Paper proposes that all gTLDs, including ".com," shall be open to registrar competition on a shared basis. Third, the Green Paper proposes that a limited number (five) of new gTLDs be established by September 30, 1998, as a transitional measure, in order to avoid "destabilizing" the Internet.

CPSR agrees with the first two of these proposals. As to the short-term creation of new gTLDs, we question whether the need for short-run competition outweighs the potential cost -- in terms of legal exposure, commercial disputes and the political constituency the Green Paper will enjoy internationally -- of the US government unilaterally dictating the opening of new gTLDs. Not only are there a number of existing entrepreneurs that are attempting to develop commercially viable registries to compete with NSI, but, as CPSR observed, there is a serious question whether the US government has the legal authority to mandate any specific domain name registration processes for gTLDs in light of the global, supra-national nature of the Internet. Moreover, except for the technical complexity of shared registries, there does not appear to be any serious technical reason why the number of new gTLDs needs to be restricted in the short run. Thus, the Green Paper's limitation of five new gTLDs is, in some respects, designed more to protect trademark holders from the difficulties of "policing a large number of top-level domains" than to respond to any legitimate issues affecting the Internet's technical stability.

While CPSR agrees that NSI should be subject to real competition in both registry and registration services as soon as is technically feasible, the importance of a bottom-up process, based on valid technical considerations, must take priority. NTIA should therefore consider whether, at least in the absence of broad international endorsement, it would not be preferable for the US government to create the new corporation, based on the principles of competition, openness and technical stability, and leave it to the new corporation to determine whether, and if so to what extent, limits on gTLD creation or introduction should be imposed. At the very least, the NTIA and the new corporation should avoid selecting new gTLDs based on competing claims of business interests of potential registrars. Names like ".web," ".pers," ".mall," etc. have been "claimed" by private entrepreneurs, and are therefore suspect.

CPSR is also concerned that maintaining monopoly provision of registries -- especially in the short run -- could have profoundly anti-consumer consequences. We believe that once new gTLDs are opened, there will be an unprecedented level of demand for second-level domain registrations. If all of these users are "locked" into a single registry (with attendant price gouging possibilities), without portability, because registry services remain sole-sourced for each new gTLD, the advent of registration competition will hardly have the benefits predicted. We do not necessarily believe that all gTLD registries must be operated on a non-for-profit basis, but rather that coordination of gTLDs zone files is required technically. If there is a way to ensure portability among TLDs and create market incentives in all DNS markets, that clearly is the preferable approach.


The Green Paper has done a notable job in deciphering the complex legal, technical and policy questions surrounding DNS reform. The principles articulated in the Green Paper -- along with its bottom-up, non-governmental approach to Internet governance -- are entirely appropriate to the global, decentralized Internet. The proposals can be improved significantly, however, by (1) entirely separating DNS administration and trademark issues; (2) creating more user-based and international representation on the "new corporation;" and (3) avoiding unilateral, US-

imposed limits on new gTLDs while opening the DNS process to real competition in both registry and registration services as soon as is technically feasible

Respectfully submitted,


By: /s/ Glenn B. Manishin____________
Aki Namioka, President Glenn B. Manishin
Harry Hochheiser Blumenfeld & Cohen - Technology Law Group
Andy Oram <>
Computer Professionals for Social 1615 M Street, N.W., Suite 700
Responsibility Washington, D.C. 20036
<> 202.955.6300
P.O. Box 717
Palo Alto, CA 94302 Counsel for CPSR

Dated: March 23, 1998


From: Sheila O'Neill <>
To: "''" <>
Date: 3/23/98 5:53pm
Subject: Response to Proposal to Improve Technical Management of Internet Names and Addresses

March 23, 1998

Re: Response to Proposal to Improve Technical Management of Internet
Names and Addresses

Dear Mr. Magaziner:

In response to the Notice of Proposed Rule Making published in the
Federal Register of February 20, 1998, the Information Technology
Association of America (ITAA) respectfully submits the following

ITAA appreciates the efforts you, officials from the Department of
Commerce as well as from other agencies have taken to construct a
transition plan to improve the technical management of the Internet,
including management of Internet domain names, Internet protocol
addresses, and the operation of root servers. These issues are central
to the continuing health and accessibility of the Internet. ITAA
supports the reliance on market mechanisms to the greatest extent
possible in the administration of Internet names and addresses and
overall, we support the U.S. Government's efforts to move to a more
private, competitive system of administering Internet names and

ITAA represents over 10,000 direct and affiliate members, including
commercial providers and users of Internet services, all of whom depend
on reliable, competitive, fair and accountable administration of core
Internet resources including domain names. A secure, reliable,
commercial grade Internet is a prerequisite for continued growth of
electronic commerce.
ITAA has closely followed the debate surrounding the reform of Internet
domain names. In the past, we have worked to provide our diverse
membership with information on domain name reform, and to facilitate a
discourse among the many stakeholders involved in the debate. In July
1997, ITAA, together with the Center for Democracy and Technology and
the Interactive Services Association, co-hosted a Forum on Internet
Domain Names. Given the complexity of the issue and the diversity of
ITAA's membership, the following comments do not necessarily represent
the views of all ITAA members. We fully expect that many of our
members will submit their own comments, outlining their individual
comments in more detail.

Consistent with the Administration's Global Framework for Electronic
Commerce, ITAA is committed to global private sector leadership in the
burgeoning field of electronic commerce. Any contemplated role for the
U.S. government , any other national government, or intergovernmental
organization should not impinge upon or prevent the private sector from
assuming this leadership role. Although it may benefit from
governmental/intergovernmental cooperation and support, responsible
Internet governance need not require governmental or intergovernmental

Competition Among Registrars and Registries
Broadly, we support the proposal to introduce competition among both
registrars and registries. Allowing an unlimited number of registrars
(assuming it is based on a shared registry, all registrars are treated
on a nondiscriminatory basis with respect to pricing, access and rules,
and that all registrars meet technical, managerial and site
requirements) will benefit users.

While a market based registry system will favor efficiency and
innovation, we are concerned about the negative effects it may have on
portability. Until registries are able to administer domain names in
more than one top level domain, domain name holders may be subject to
"lock-in." Domain name holders will have a disincentive to relinquish
an existing TLD because of the negative business implications and the
costs associated with establishing a new TLD. We are also concerned
that the initial five new registries will have a considerable market
advantage over potential new registries who may wish to enter the market
after the transition period.

In the spirit of true competition, we believe that both commercial as
well as non-profit entities should be eligible to serve as registries,
as long as they meet the highest standard of technical, operation,
managerial and other criteria.

Coordinated Functions: The New Corporation
ITAA supports the creation of a private, not-for-profit corporation (new
corporation) to oversee the policy guidance and control of top level
domains (TLDs) and the Internet root server system. We would caution
against prematurely determining the number of seats to be assigned to
the numerous stakeholders in the Internet community. The creation of
yet another membership association to represent Internet all users
(i.e., commercial and non-commercial) may prove impossible. ITAA
believes there already exist sufficient numbers of membership
associations in the U.S. that represent commercial Internet users. Also
the composition of the new corporation must be flexible enough to
readily reflect the diverse interests, both commercial and
non-commercial, of the Internet community. We agree that officials of
governments or intergovernmental organizations should not serve on the
board of the new corporation.

While we support the creation of the corporation under U.S. laws because
it will provide our members with an acceptable degree of certainty, in
the long term it may be necessary to provide for a more
internationalized governance mechanism. ITAA is committed to the
principle that the Internet is a truly global medium.

ITAA members include companies responsible for the building of the
Internet infrastructure as well as companies that conduct a growing
percentage of business over the Internet. As such, we are equally
concerned with both the operational as well as trademark aspects of the
management of Internet names and addresses.
The Internet today is a commercial medium. If the promise of global
electronic commerce is to be fulfilled, there must be effective policing
and protection of trademarks. Domain names in fact have become a form
of brand names. This includes protection from cybersquatting.

While in the long run we support the introduction of several new TLDs,
we have serious reservations about the immediate introduction of 5 new
TLDs because it may cause more problems for trademark holders.
Nevertheless, we should proceed with caution to create new domain names,
especially to meet the demand from business and industry sectors and not
to restrict the growth of the Internet and electronic commerce. In
parallel, we support continued efforts : 1) to establish precise
standards and rules for trademark-like protection for domain names, 2)
to create international searchable databases to secure effective
enforcement, and 3) to harmonize internationally trademark laws. We
believe that technological advances, including the creation of
international searchable databases, will eventually provide an effective
means of avoiding domain name and trademark disputes. In the long run,
the international harmonization of trademark law will provide business a
stable commercial environment with clear definitions and redress to
trademark infringements.

We commend you for your continued consultation with Internet
stakeholders and look forward to working with you on this important

Harris N. Miller

CC: "'domain names:'" <domain names:...


From: Mike McDonnell <>
To: NTIADC40.NTIAHQ40(dns)
Date: 3/23/98 10:11am
Subject: comment

Domain Name Naming: Free Speech, SPAM, and XXX Adult Sites

In talking with various people at Internet World about my idea for making life easier for web users and safer for children to use the Internet is to have simple, obvious domain extensions.
.com is commercial
.ads is for advertising
.adt is for adult content

Since the US is the major player on the Internet it seems the US does have the power and responsiblity to make the Internet a safe and sane place for all.

Mike McDonnell


From: KathrynKL <>
To: NTIADC40.NTIAHQ40(dns)
Date: 3/23/98 6:02pm
Subject: Trademark/Corporation Comments from DNRC and NetAction

Attached please find DNRC and NetAction's Comments to the Green Paper.
This set of comments addresses issues of of trademark law, jurisdiction,
openness of process, privacy and balanced representation within the
Corporation's Board of Directors.

The document is formatted in Word 6.0/95. If you have an difficulty reading
this file, please let me know.

Kathryn Kleiman
General Counsel, A-TCPIP/Domain Name Rights Coalition

[View attached file]


From: Jay Fenello <>
To: "Roeland M.J. Meyer" <rmeyer@MHSC.COM>
Date: 3/23/98 5:00pm
Subject: Domain Marks (was: domain police)

At 06:33 PM 3/22/98 -0800, Roeland M.J. Meyer wrote:
>>Doesn't the GP only state that the Registry and Registrars are to request
>>certification from the applicant that the requested name does not infringe
>>the Trade Mark Rights of another?
>>SAme with Company names and Business name sin AU (as well as Domain Names).
>> It's the appplicants reponsibility to ensure no conflict.
>>That's how we operate.
>But, what do you do if a conflict is presented? One can do a "due
>diligence" trademark search but to do so in the international business
>community is virtualy impossible. If it were an absolute requirement for
>internet business than we can shut the whole mess down right now and save
>ourselves the grief. Someone, somwhere, can claim a apriori restraint on
>virtually *any* name. Ergo, no new domains and a radical reduction in
>existing domains. This is not good.
>We need some common-sense, lacking that, we need to keep the TM rascals out
>of the domain name business.

Iperdome addresses this issue through the concept
of a Domain Mark(sm) as part of our response to the
Green Paper. Please see:


Jay Fenello
President, Iperdome, Inc.



From: Jay Fenello <>
Date: 3/23/98 4:41pm
Subject: Iperdome's Response to the Green Paper

Iperdome's response to the Green Paper
has been submitted, and is available
for viewing at:


Jay Fenello
President, Iperdome, Inc.


From: KathrynKL <>
To: NTIADC40.NTIAHQ40(dns)
Date: 3/23/98 6:06pm
Subject: Free Speech Comments from DNRC, NetAction, Prof. Mueller

Attached are comments submitted jointly by DNRC, NetAction and Professor
Milton Mueller of Syracuse University and signed by 215 individuals, community
groups and Internet-based organizations. These comments urge the White House
and Department of Commerce to add to the governing principles of the new
Corporation a principle in support of free speech and open communication.

The document is formatted in Word 6.0/95. If you have an difficulty reading
this file, please let me know.

Kathryn Kleiman
General Counsel, A-TCPIP/Domain Name Rights Coalition

[View attached file]


From: <>
To: NTIADC40.NTIAHQ40(dns)
Date: 3/23/98 12:13pm
Subject: The .COM Owners Prayer

Now I lay me down to sleep, I pray the Lord my domain to keep. For
if I should wake and find NSI has screwed up my domain (again), I
pray the Lord my soul to take.

This is the prayer I say at night and so far it is keeping Network
Solutions from shutting off my domain by mistake again. I recommend
to all those who have suffered inexcusably poor service from Network
Solutions to say this prayer every night before you go to sleep.

Maybe God will bless the world and give Ira a vision of a future
without Network Solutions in our lives. That would be as close to
Heaven on Earth as one could get....

God Bless You Ira!
Nicholas Bentworth

Get your free vanity email address at


From: Einar Stefferud <>
To: NTIADC40.NTIAHQ40(dns)
Date: 3/23/98 4:53pm
Subject: Einar Stefferud Comments on Green Paper

Before the
National Telecommunications and Information Administration
Washington, DC 20230

In the Matter of )

Notice of Proposed Rulemaking

Comments of Einar Stefferud


Einar Stefferud, President
Network Management Associates, Inc.
17301 Drey Lane
Huntington Beach, CA 92647-5615


23 Mar 1998

-------------------------------- break --------------------------------


Summary [1-7]
Qualifications of Einar Stefferud [8-17]

I. The Proposal
I-A. The Coordinated Functions [18-31]
I-B. The Competitive Functions [32-38]
I-C. The Creation of New gTLDs [39-45]
I-D. The Trademark Dilemma (ref. App. 2) [46-50]
I-E. The Intellectual Infrastructure Fund [51-53]

II. The Transition
II-A. The NSI Agreement [54-54]
II-B. Competitive Registries, Registrars, and [55-59]
the Addition of New gTLDs (ref. App 1)
II-C. The Root Server System [55-59]
II-D. The .us Domain [65-69]
II-E. The Process [70-75]


1. The U.S. Government efforts to restore sanity and due process to
the matter of coordinating and managing the Internet Domain Name
System (DNS) are very much appreciated, especially in terms of how
this action supports cooperative efforts within the Global Internet
Community to employ traditionally open Internet "Rough Consensus and
Running Code" development processes. This is in contrast to what has
recently become a fierce "Zero Sum Game" wherein many separate groups
have been fighting among themselves and lobbying various authorities
and governmental institutions in an effort to become the "winner" of
the right to control the DNS. That way there be dragons!

2. It is my very strongly held opinion that the Internet and its DNS
does not consist of a "zero sum game" situation wherein there is a
naturally limited supply of TLD and SLD value that must be divided
among a growing set of players. Instead, the Internet and its DNS
should be seen as a vast opportunity to create new value for the
benefit of all manner of participants. Fighting over some arbitrarily
limited number of DNS names does not make any sense. The current
arbitrarily small number (3) of openly available non-national TLDs
appears to be the primary cause of all the present anguish and anger.

3. I am also pleased to note that this U.S. Government effort is
focused on removing itself from its unilateral controlling position
with a careful transition to a broad global system of coordination and
management. The Internet is not confined to any single nation, and I
believe its Domain Naming System must not be controlled by any single
nation or single collection of nations.

4. Indeed, it is my experienced judgement that the Internet should
not be controlled by any Inter-Governmental agency in the manner of a
cartel such as has been employed for telephone, telegraph and radio
spectrum management. The DNS namespace is not a naturally limited
space, any more than our language is naturally limited in terms if
finding and using new words with new meanings. The history of
governmental attempts to control the use of words in national
languages does not encourage any confidence that DNS names can or
should be so controlled.

5. The Internet is a complex web of cooperating private networks that
only works as it does because of the absolute need for all involved
parties to cooperate with each other in order to use the Internet.
The Internet has been built with this kind of cooperation, and without
it, the Internet can only wither and die. To let it wither and die
because central controls must be applied is not an acceptable outcome
in any terms that I can imagine.

6. Thus it is critical for the next phase of evolution of the
Internet for it to continue building on the fundamental structure of
mutual cooperation for mutual benefit, and to do so with a minimum of
disruption during the proposed transition from U.S. Government control
of certain administrative functions (IP Number Assignments, DNS, etc.)
as now provided by IANA (Internet Assigned Numbers Authority).

7. My comments in this submission are largely focused on issues
related to the Domain Name System (DNS) and only peripherally related
to other aspects of IANA such as IP Numbers. Of course, how the
required coordination of the DNS Root Service fits into the larger
context of overall Internet coordination and management is among the
issues of how the DNS Root Service is coordinated and managed.

-------------------------------- break --------------------------------

Before the
National Telecommunications and Information Administration
Washington, DC 20230

In the Matter of )

Notice of Proposed Rulemaking

Comments of Einar Stefferud

Qualifications of Einar Stefferud [8-17]

8. Einar Stefferud respectfully submits comments in this proceeding,
as a semi-retired independent consultant with 34 years experience in
the field of Shared Computer Resource Management with focus on the
governance of networks which invariably accompany efforts to share
resources. The Internet is the penultimate instantiation of resource
sharing. Mr. Stefferud has focused his work since 1971 on strategic
management and governance of internet environments. DNS governance
issues addressed in this rule making process are central to Internet

9. Most recently, during this past year, Mr. Stefferud has been very
much involved in cooperation with an ad hoc group of variously
talented and experienced Internet Domain Name System experts in an
attempt to build consensus for an Open Root Service Confederation to
provide for coordination of comprehensive DNS Root Service operation.
10. The results of this work-in-progress can be found at
<>, and by reference this work is hereby
included as an integral part of these comments. They represent the
result of our collective cooperative thinking, while the comments
here-in reflect only my own personal and professional experience.

11. Before 1970, Mr. Stefferud served (in reverse order): On the
Corporate Planning Staff of System Development Corporation; as Manager
of the Carnegie Institute of Technology Computation Center; as a RAND
Corporation consultant; and as a Graduate Student Assistant in the
Graduate School of Management at UCLA while earning BS & MBA Degrees.

12. From 1975 until 1986, Mr. Stefferud served as Moderator of one of
the first ARPANET Mailing Lists (MsgGroup) which discussed Electronic
Mail issues. Archives are available at <>.

13. Between 1980 and 1995, he served as a visiting instructor or
adjunct professor at the University of California at Irvine, teaching
electronic mail, computer networks, and social impacts of computing.

14. He founded Einar Stefferud and Associates in 1969, which was
incorporated as Network Management Associates, Inc. (NMA) in 1973. To
the present time he serves as President of NMA, located in Huntington
Beach, California. Services provided by NMA include working with all
levels of management to deal with the full range of policy,
governance, planning, organization, architecture, implementation, and
operating issues relating to Internet environments.

15. NMA clients have included Harvard University; First Virtual
Holdings; Microsoft; MITRE Corp.; Allied-Signal Aerospace; Deloitte &
Touche; IPnet-Solutions; Freeport-McMoRan; Isocor; Infonet;
Soft-Switch; TRW; NASA Ames Research Center; US Army Materiel Command
(including ARDEC, ARADCOM, BRL, AVSCOM, and TECOM); US Army Computer
Systems Command; FCC; Northrop; Wollongong Group; Digital Equipment;
SRI International; Univ. of Arizona; Univ. of California at Irvine;
Univ. of Delaware; Univ. of Kansas; Univ. of New Mexico; Temple Univ.;
the Univ. of Wisconsin; and Brigham Young Univ. & the LDS Church.

16. In 1994, he co-founded First Virtual Holdings Incorporated, with
Dr. Marshall T. Rose, Dr. Nathaniel Borenstein, and Lee Stein, JD. In
1994, First Virtual mounted the first Internet payment system.

17. He was honored as one of the Top 10 Visionaries in the Computer-
Communications Industry for 1993 by Communications Week Magazine, for
his contributions to the Internet. (Communications Week, 23 Aug 1993)

-------------------------------- break --------------------------------

I. The Proposal

I-A. The Coordinated Functions [18-31]


18. The Internet needs to be sheltered from both threats of rampant
anarchy and from threats of imposed central authority from would-be
dictators, be they posed by individuals or institutions. Until
recently, IANA provided shelter from both of these threats. However,
the Internet has shown itself to behave as a "region of bounded chaos"
in which rampant "speciation" continues to yield a rich array of new
products and services with powerful network effects and economic

19. The current IANA was structured as a simple Czar style authority,
as is generally used for coordinated management of all wholly owned
networks, such as a telephone company's network, or what has recently
come to be called an INTRAnet, wherein all components in a network are
owned and controlled by a single person or organization. IANA was
born out of the original ARPAnet which was wholly owned by the
U.S. Government and operated by U.S. Government contractors. This
originally established a tradition of central IANA czarist-style
control with U.S. Government oversight. For many years, we all fondly
referred to Jon Postel and IANA as "the Internet Numbers Czar". This
was a genuine term of endearment, and it remains in place to this day.

20. With creation of the NSFnet and then the Internet as we now know
it, something new was created. Namely, a network of networks, not all
owned by any single entity, so that it could only work with end-to-end
cooperating owners and operators of the aggregated network of private
networks. Thus, gradually over time, the rationale for a czar-like
top down central authority disappeared with Internet growth, but IANA
did not adjust to this new reality. Thus, in due course, IANA has
become disfunctional as the czarist coordination agent. The Internet
demands more open consensus based coordination, just because there is
no singular source of authority to be used for purposes of control.

21. Thus, IANA must either be reformed into a confederation of
customers and providers of DNS and IP Assignment Services, or some
externally governed agency must impose its will on the Internet
through a takeover of these IANA coordination functions. Clearly I
prefer an open confederation, styled after the structure of a self
governing customer or producer cooperative, which will make the
administrators and operators answerable to the broad community of
Internet Service Providers, local network managers, registries,
registrars, server administrators and users.

22. In any case, to properly deal with the open nature of the
Internet, any replacement or reorganization of IANA must also be open
in terms of its operational processes, the selection of its management
and governing boards, and its decision processes. IANA has too long
been operated in a closed manner.

23. In my considered opinion, the NPRM proposal, as it stands, does
not completely fulfill the need for open confederated coordination
mangement. As described, it appears to be a top down structure, with
a Board of Directors that is supposed to be representative of the mass
of internet service providers and users, but for which the method of
selection of representatives is very vague with definite possibilities
of loading the board with serious conflict of interest problems, or
with people who simply do not understand the Internet and its needs.


24. Board of Director seats must be allocated with consideration
given to voting power within the board to avoid granting any given
group control by virtue of having 50% or the seats. This means that
the seats must be allocated not in terms of percentages of the total
number of seats, but in terms of percentages of the number of votes
required to control decisions. To wit: 51% of the vote equates to
100% of control. So, to give any group 50% of control, they should be
granted only 25% of the vote. This will assure that any given
represented group must convince a significant number of members from
other groups to go along any proposed decision. I suggest that no
identifiable group should ever be given more than 25% of the vote,
especially so called "users".

25. It must be recognized that the TLD registry operators and owners,
taken together as a group, must have a large voice in how coordination
is to be achieved, politically, technically, and operationally. So
should their associated Registrars who will have a different but
clearly related view of things. IP Number assignment registration
groups certainly deserve equal representation with the registries and
registrars, and there must be some way to represent what we like to
call "users" and also what we like to call "customers".

26. Users and Customer representation poses very interesting issues
in terms of their particular value judgement processes. I find it
very informative to distinguish "customers" from "users". Both tend
to say the same words, but mostly with very different meaning. When a
customer says "I want X" they are making an offer to buy X, and have
made a value judgement that X is worth its price. When a user says "I
want X" it only means that they want X, with no offer to buy attached,
and no value judgement implied. Typically, in the workplace, the
users use what their employers provide, and when given more will "use"
it, and often ask for more. Their employers, acting as customers,
must evaluate whether or not providing more is worth the result.
Hence, customers make value judgements, and "users" do not. Users
tend to be infinite sinks for free resources. So, "users" run up the
bill, and "customers" decide whether or not to pay for it. Customers
have budget responsibilities, and "users" as distinguished here, do

27. It is my firm recommendation that "users" as described here not
be given any seats on the Board of Directors! Customers YES, Users
NO! I have seen universities form Computer Committees with half and
half, "users" and "customers", resulting in total deadlock between the
two camps preventing them from making decisions. This result must not
be allowed to occur for the Internet.

28. An alternative scheme for selecting members is to encourage
formation of open confederations of registries, registrars, local DNS
resolver administrators, and DNS root server operators, and let each
of these select members of a DNS Board of Directors to deal with DNS
issues, and select delegates to the NPRM Board of Directors. The IP
Number Assignment organizations should likewise organize to manage
their own coordination and select delegates to the NPRM Board.
Internet Service Providers also need a clear voice in the NPRM Board.
Thus the entire governing structure can be made to work very much like
a cascade of customer/producer cooperatives in a proper bottom up
power aggregation arrangement. The key here is "power aggregation" in
place of "authority delegation".

29. This is in contrast to the concept in the NPRM of attempting to
name proper representatives to the top level NPRM Board which would
then in turn act to delegate authority and assign responsibilities
down to the DNS ROOT operators and the IP Number Assigners in a
traditional top down control control mode of operation. At the end of
every day, this arrangement depends on always having selected the
right members of the top level board, because it will be given
authority to run things as it may see fit. This puts us right back
into the czar-like model that we are trying to replace.

30. In particular, the NPRM already suggests specific existing boards
to select delegates to the NPRM, such as the Internet Architecture
Board (IAB). It should be noted that the IAB is not a particularly
representative board, and is under some serious criticism for its
apparent biased support of the IAHC/POC/CORE efforts to capture sole
control of the DNS root. It would not build much confidence in the
Internet Community to allow them to unilaterally appoint 1/7th of the
members of the NPRM Board. If the IETF community should be
represented on the NPRM Board, they should be selected directly by the
IETF Nominating Committee (NOMCOM) which selects the members of the
IAB, and also selects the Area Directors of the IETF which serve as
members of the Internet Engineering Steering Group (IESG). The NOMCOM
will provide a much more reliable source of proper representation of
IETF interests, drawn directly from the IETF aggregated power base at
the Working Group level.

31. The reality is that the IAB is not a good representative of the
IETF community, because it actually serves more as a subcommittee of
the ISOC than an executive reporting point for the IETF. Neither the
ISOC or the IAB have a good record of representing a broad segment of
the Internet community at large. Both have very narrow purviews and
limited vision in matters of management and governance. And, both
tend to conduct their decision making in closed private sessions.
Both have been doggedly supporting the IAHC/POC/CORE program in spite
of broad opposition. This discredits their claims of representation.

-------------------------------- break --------------------------------

I-B. The Competitive Functions [3-38]


32. Open competition is the best cure for most of what ails the
current DNS situation, and this should be applied to support
competition among both registries and registrars. But it is not at
all clear that every TLD should be structured like every other TLD, as
proposed in the NPRM. There is no reason to believe that every TLD
will have the exact same purpose for its existence, so a requirement
that each and every TLD be structured in the exactly the same way to
avoid its being considered a monoploy fails the test of rationality.

33. For example, one experimental TLD (.ZOO) is actually organized to
support zoological taxonomies, and as such should be fully controlled
by a management board properly representing the field of Zoology. For
what reason would it be helpful to force this TLD to allow any and all
SLD Registrars to have privileged access to register names on a first
come first served basis?

34. Another example might by the Dun and Bradstreet number
registration system, which might want to mount their entire number
registry under .DUNS. Or the New York Stock Exchange might want to
place all their stock registration symbols under .NYSE? and the same
for .NASDAQ. What is wrong with allowing these things to happen? I
see nothing at all wrong with allowing it, as they all offer great
potential economic value.

35. Consider .IBM (if if IBM ever wants to claim it) for internal IBM
sub-domains? Of what use is it to require that IBM divisions be
allowed (or forced) to go through external free lance registrars to
register in their own company's TLD? It is simply very premature to
take such a highly restrictive approach to forcing competition,
especially where it would be totally inappropriate.

36. Indeed, it is very easy to imagine many globally trademarked
companies wanting their own TLD, and there is no reason what-so-ever
to force them to use external registrars for internal registrations.
In fact, this might be a good way to help with the Trademark Dilemma,
by allowing Global Famous Trademarks to become TLDs.

37. Further, a free ranging SLD registrar industry has already formed
to deal with the many different nTLD and iTLD registries on behalf of
registrants who want to take advantage of their knowledge of where to
get the best deal, and how to deal with the different policies and
processes that exist among nTLD registries. This registrar industry
is already putting pressure on NSI and other registries to deal with
registrars as a retail distribution channel for their products.

38. It seems fair that NSI be required, as part of the
U.S. Government process of negotiating the "wind down extension" for
the NSI Cooperative Agreement, to operate as proposed in the NPRM.
NSI currently enjoys a healthy monoploy on .COM, and they have a
history of resisting price discounts for registrars. But, I see no
special reason to burden every other TLD registry with requirements
that are specifically tailored to curb NSI monoploy practices when all
other TLD registries are essentially in a start up condition.
Hobbling all non-NSI registries will only serve to prolong NSI's
insulation from open market competition.

-------------------------------- break --------------------------------

I-C. The Creation of New gTLDs [39-45]


39. Actually I prefer the term "iTLD" standing for International TLD,
just because gTLD is a term devised by IAHC/POC/CORE to convey the
notion that Generic TLDs are part of some kind of public trust, which
is a concept the I totally reject. Hence the term is distasteful.

40. I applaud the concept of conducting some experiments with
insertion of some relatively small number of new iTLDs, but I disagree
about arbitrarily setting a very small fixed limit of 5 new iTLDs.
The primary issue with needing to experiment is much more a matter of
the rate of infusion rather than the total number of infusions. It is
not easy to translate a fixed upper limit into a rate. So, I propose,
in concert with my colleagues in the Open-RSC <> that
the U.S. Government simply adopt a rule that new iTLDs only be added
one at a time, while monitoring the stability of the root and of the
DNS as a whole, and monitoring the impact of new iTLDs on the
Trademark situation, etc.

41. There exist more than 5 experimental iTLD registries that were
mounted a long time ago, and should be given fair treatment in terms
of their insertion into the DNS ROOT. By declaring a small arbitrary
limit on the number new experimental insertions, the situation is
immediately reset to continue the original zero-sum game, since anyone
that gets into the first 5 must have displaced some other candidate,
and this forces the U.S. Government, or its new agents, to play
Solomon and make very wise selections from among too many candidates.
I propose that the limit be set by counting the number of candidates
that pass muster as of some certain date, such as 1 May 1998.

42. It will work much better to set objective criteria for queuing
the insertion of new iTLD names into the root, in serial order, with
no more than one insertion at a time. In this situation, the pressure
to be among the first 5 will disappear, as there will be no reason to
fear being cut off for arbitrary reasons. Indeed, the queued
candidates might be expected to amicably negotiate their relative
positions, given that they would not be giving up their insertion
rights altogether!

43. Further, in order to facilitate testing and qualification of each
new iTLD, a "TEST ROOT" service should be mounted, separate from the
"AUTHORITATIVE ROOT" where candidates can be entered for their own
testing, and for a qualification test environment where they can be
thoroughly tested before being added to the AUTHORITATIVE ROOT (now
known as the IANA ROOT).

44. Such a test root already exists as a Root Service Confederation
with a significant set of candidate iTLD registries already in
operation, most of which are already in compliance with the Green
Paper proposed criteria for new iTLD additions. What needs to be done
next is establish criteria for admission to the IANA ROOT and proceed
with conformance testing.

45. So, to sum up, it is clear to me that the setting of arbitrary
small limits on the number of new iTLDs to be added is a main cause of
much of the fractious infighting, posturing and positioning of
expectant iTLD registries, and that setting a process of single file
insertion, at a rate geared to maintain control and stability of the
root and the DNS, is a vastly better way to proceed, as it greatly
reduces the anxieties and the risks that force the community into
dealing with the situation in zero-sum game terms.

-------------------------------- break --------------------------------

I-D. The Trademark Dilemma (ref. App. 2) [46-50]


46. It is very clear to me that limitations or reductions in the
number of iTLD names in the root will not solve any trademark dilemma
issues. It is already the case that nTLDs are acting just like iTLDS,
by accepting all manner of registrations from anyone anywhere on the
globe, for any reason, including commercial use. There is already a
brisk business in registrations under .TO and .NU, and .TM has been
reserved by its Nation Owner for "International Trademark
Registrations" for very high fees. It is not going to be enough to
just control the number of iTLD names in the root.

47. The reality of all this trademark argument is that the same
fights over the use of trademarked names will occur with or without
new iTLDs. The only thing that will change is the level in the DNS
tree where naming conflicts are perceived. If all commercial users of
the net must register in .COM, then the focus will be on .COM SLDs.
So, by restricting the TLD level, the problem moves down the tree, and
by opening up the TLD level, it merely moves up the tree.

48. The basic problem is caused by the Trademark Industry's
traditions, lodged now in many national laws, that allows use of same
name by different "owners" in different market segments, even in the
same geographic territory. The DNS does not allow any two different
owners of the same fully qualified DNS name anywhere in the DNS tree,
so actually, the DNS system is much more stringent than Trademark Law
in regard to shared use of any single trade name.

49. Thus, the solution is going to have to be devised by the
Trademark Industry and not be the Internet DNS community. Restricting
the DNS to serve the Trademark Industry is pointless, as noted above,
because all we can do is massage the problem up and down the DNS tree.
They cannot make it go away because the problem is that the trademark
holders perceive infringement wherever they please to see it, and they
certainly please to see it at the highest possible DNS level. Pushing
the problem down the tree by restricting the number of TLD names will
also push their perception down the same tree, with no reduction in

50. So, it seems clear that the long term solution requires time for
the courts to come to grips with the fact that DNS use of a
trademarked name is not necessarily an infringement of that name.

-------------------------------- break --------------------------------

I-E. The Intellectual Infrastructure Fund [51-53]


51. It seems clear to me that the funds in the Intellectual
Infrastructure Fund (IIF) have been paid into it by Internet Community
participants, and that if that money belongs to anyone, it belongs to
the Internet Community.

52. I propose that the funds appropriated by the US Congress be
returned to the IIF, and that the grand total of the funds collected
be placed in an Internet Trust Fund which will operate to fund
proposals for the use of the funds for the good of the Internet.

53. Among the first proposals I would expect to be submitted would be
those related to funding the establishment of the new coordination
mangement organizations that are proposed in this NPRM. But, I would
buffer them from manging the IIF by requiring that they apply for
grants for specific purposes. In general, the new coordination
management should organize themselves to be self funding from their
supporting confederations.

-------------------------------- break --------------------------------

II. The Transition

II-A. The NSI Agreement [54-54]

54. No comment. The plan looks fine to me.

II-B. Competitive Registries, Registrars, [55-59]
and the Addition of New gTLDs (ref. App 1)

55. As noted previously, open competition among both TLD registries
and Registrars is a good thing and will serve to solve most of the
problems faced by the DNS.

56. But, the cure must not be overdone, as noted above, by requiring
open competition in every instance because we already know that there
will be instances of new TLDs that are owned and used by companies
only for internal registrations or possibly for special cases such as
for zoological classification systems and private ID numbering systems
like DUNS or NASDAQ or the NYSE.

57. It is also unreasonable to burden new startup TLD registries with
all the costs required to emulate the cure being applied to NSI.
Further, it is not required for startups, which will have to compete.
58. In any case, the Registrar industry is already quite well formed,

and it will be natural for new TLD registries to arrange for retail
Registrar resale of of wholesale TLD Registrations. The Industry is
already in operation, with a variety of flavors of Retail Registrars.
These include many ISPs that bundle registration with access services,
and pure Registrar operations that specialize in knowing how to deal
with a wide variety of registries.

59. And, as noted previously, in transition, it will be vastly better
to permit more than 5 new iTLDs to be added, so as to avoid all the
stress and strain of deciding which 5 out of 10-20 candidates will be
added. Better to serial order all qualified candidates according to
objective criteria such as date of first operation, and continuity of
operation since then, and readiness for insertion into the root by
virtue of meeting all the operational requirements which must also be
stated and evaluated objectively.

-------------------------------- break --------------------------------

II-C. The Root Server System [60-64]


60. IANA plus the U.S. Government plus IAB plus NSI is not a broad
enough cross section of the community to include a enough experienced
people in the proposed NPRM study.

61. In particular, it does not include any of the people who have
already been running Root Server Confederations for several years.
Thus the study will be heavily biased against the ideas put forward by
the experienced open root service confederation operators.

62. Further, it is well known (from reading the IETF and other
mailing lists), that IANA and the IAB have generally regarded the
people who are successfully running root service confederations to be
an unsavory lot. So, the study as proposed looks like a matter of
sending the foxes to guard the chicken coop.

63. It is really important for all concerned to understand the
dynamics of a cooperative confederation of root server operators who
are all bound by mutual needs to include each others data in their own
servers, and thus guarantee comprehensiveness and coherency of the
collective confederation root service. This appears to be more easily
done with a bottom up confederation than with top down authority.

64. I suggest that the U.S. Government take a close look at the work
in progress by the Open Root Confederation at <> to
see how such an organization can be put together without governmental
funding or governmental involvement.

-------------------------------- break --------------------------------

II-D. The .us Domain [65-69]


65. Cyberspace is not Geospace, and mapping one onto the other is a
pointless concept when offering domain names for sale. It locks
people into staying where were located when they registered, or it
means that they move without changing their names, thus defeating the
whole idea of enforcing geo-locality relationships. The result has
been a large number of unpalatable names forced on prospective
registrants who would much rather obtain a less than choice name in
.COM, or even in another country's nTLd.

66. The .us TLD should be given over to a more accommodating
administration that does not believe in its God Given Authority to
impose silly geo-space concepts on hapless victims.

67. There is absolutely not reason why the IANA, with all its
difficulties with managing the International DNS situation should be
saddled with also managing the domestic .US TLD.

68. One possibility would be to auction the .US TLD off to a
commercial operator, or have the U.S. Congress commission a non-profit
company, or assign it to the U.S. Postal Service, or ask the States to
cooperatively organize an operating corporation to run it, or some
such. In any case, it should be given over to someone with
operational experience in delivering services to the public, and with
some real marketing experience. The USPS has recently shown that it
has learned some important marketing lessons.

69. In short, the .US results show that IANA has no sense of
marketing because all the .US problems look like failed marketing to
me. It certainly is not a technical problem.

-------------------------------- break --------------------------------

II-E. The Process [70-75]

70. If it is the goal of he U.S. Government to gracefully exit as
fast as possible, leaving behind a solid consensus agreement on the
structure it puts in place, it must spend more effort to support
consensus generation in place of the past and present zero-sum game
squabbling over the spoils.

71. This means that the U.S. Government needs to support real open
deliberative processes and support those people who are already
successfully working toward building consensus.

72. To make this happen, the U.S. Government must make it very clear
to all concerned that it is not going to grant special favors to any
group as a reward for mounting any massive lobbying blitz, or by
threatening and bullying anyone, or by engaging in personal attacks
and uncivil discourse in open discussions. It is by holding open the
political possibility that heavy lobbying or blistering blustering
attacks on others will result in the perpetrators gaining the upper
hand in the DNS wars, that the U.S. Government blocks cooperation and
encourages fighting.

73. This is the root of the entire problem, and it will remain the
root of the problem as long as there remains any hope for anyone that
the game is to fight and win over the losing players, rather than to
cooperate and win/win.

74. In this regard, I commend the work of the Open-RSC
(> as an example of what can be done with some simple
civil discourse rules for mailing lists, and a serious attitude about
remaining open to the public on the mailing list no matter what issues
arise there.

75. It is interesting to note that even though the Open-RSC group has
no formal contract bond, and is just a variably talented group of
people, including some that are considered impossible to work with, it
has not yet had to invoke its very simple "Civil Discourse" rules or
penalties. And, it has achieved a large measure of consensus on how
to organize to manage a confederated open Root Service. It is now
running such a service on an experimental basis.

---------------- END OF EINAR STEFFERUD NPRN COMENTS -------------



From: Eric Weisberg <>
To: Mike Nelson <>
Date: 3/23/98 5:09pm
Subject: Comments of Internet Texoma, et al in Docket No. 980212036-8036-01

The following attached in RTF) are the comments of Internet Texoma,
Inc., Eric Weisberg, Larry Vaden and Peter Veeck on the Proposal for the
Improvement of Technical Management of Internet Names and Addresses.

National Telecommunications and Information Administration
Washington, DC 20230
Notice of Proposed Rule making

In the Matter of

Comments of
Internet Texoma, Inc., Eric Weisberg, Larry Vaden, and Peter Veeck
By Eric Weisberg, General Counsel
200 W. Main, Denison, TX 75020-3025
submitted March 23, 1998


We are ERIC WEISBERG, General Counsel, and LARRY VADEN, Founder and CEO
INTERNET TEXOMA, INC., filing this individually and on behalf of
Texoma, Inc., an ISP which serves North Texas from the Dallas calling
scope to the Oklahoma border. Our subscribers are primarily in rural
areas. Internet Texoma is a member of ARIN, ISP/C, and TISPA. We are
joined by Peter Veeck, owner of Regional Web, a website for local
in the areas served by Internet Texoma. We are all founders of the
American Internet Registry (NAIR), an "organization in formation" which
filed an application with the IANA for a block of Internet Numbers " for
the purpose of acting as an IP registry on a non-profit or profit making
basis (we asked for guidance on that from the IANA, but received no
response). No further action has been taken relating to that project.
have no other direct financial interest in the proceeding except as tax
payers and as stake holders in the overall success of the Internet.


A. Principles to be Applied

We move that the following principles of FAIRNESS guide the decision
process you have undertaken:

1. Representation of all significant interests,
2. Equal access to resources,
3. Competition instead of monopoly solutions where possible,
4. Due Process in adjudication, and
5. Open meetings and records.

B. Organization of Comments

Internet Texoma will discuss specific parts of the proposal in the order
they appear in the document. As will become clear, we have some
objections which transcend the specific language.

C. General Objections, Observations and Recommendations.

1. We specifically OBJECT TO the creation of PRIVATE MONOPOLIES in
name REGISTRIES unless there is strict regulation or other means of
protecting the public from unfair or predatory practices.

Private Monopolies offer no benefit over government monopoly control.
Private monopolies subject consumers to price gouging and oppression
without the real benefits of competition. We do not see domain name
registration as an inherently entrepreneurial enterprise. On the other
hand, we value the legal protections afforded the public in dealing with
governmental entities--representative governance, open meetings and
records, and due process. We oppose the loss of those protections
the benefit of completely open competition. And, we object to the
incredible windfall profits to the monopoly registries which may result
from this proposal.

2. If you attempt to create a registry system dependent upon
by the MARKET (which is a questionable proposition given your proposed
approach and the difficulty and expense of changing one's name), the
(IANA or its successor) MANAGEMENT (or the registries will have to
be contractors carrying out the technical function for others who
control the TLD, as will be discussed in item 3 below). That would
price and service leadership and competitive regulation on any private
non-profit monopoly in the new registries which are allowed.

3. Or, you will have to adopt a different approach than is proposed for
ownership or control of TLDs -- one which will ELIMINATE the issue of
MONOPOLY REGISTRIES. That might be accomplished if the registries do
have any right to the domains they register, but only perform the
function as contractors to registrars or registrar coalitions or
federations. Registry organizations should be able to manage as many
domains as they can get to contract with them. The registrars/registrar
federations should be able to change registry contractors whenever they
fit. We believe that this approach will allow competition between
registries as well as registrars. To this end, the registry
property which the USG recovers from its contractor, NSI, should be
in the public domain for the use of all registries and registrars.

4. The criticism that this proposal fails to deal with the
nature of the Internet must be considered. There must be broad
international representation on the board of the new Corporation,
through representation of regional organizations.

5. We suggest that various approaches be allowed simultaneously - CORE ,
eDNS, Iperdome, AURSC, AlterNic etc. should be allowed to participate
the new DNS world we are creating.
6. We further OBJECT to any TRANSFER of ownership of or interest in the
EXISTING TLDs to any private entity or any contract to manage the
function for existing TLDs WITHOUT COMPETITIVE BIDS. These were
at US Government expense, are of significant value, and there is no
justification for giving these resources to private parties to exploit
without full compensation to the Government.

7. We OBJECT to any PREFERENTIAL TREATMENT for the current DNS

8. We DEMAND the RETURN to the registrants or the US Government of all
moneys and other BENEFITS RECEIVED by NSI over and above the cost-plus
contract originally entered into with NSF. We believe that the 1995
amendment to the so called "Cooperative Agreement" is illegal and
and disgorgement of all monies and benefits flowing from the amendment,
including the hardware and software bought and/or developed with such
should be compelled.

Thus, we MOVE that your demand return of all such assets, immediately
refer all unresolved questions, including disgorgement, to the
of Justice to determine the USG's legal interests and how beset to

9. We question whether the proposed mechanism for electing the board to
govern the IANA's successor will result in fair representation for all
legitimate interests.

10. Since the seat of Internet governance must be somewhere and subject
some law, there is no reason that the US should not act to assure that
is located in the United States, is organized in a fair and
manner, is subject to the rules of open governance and due process as
be applicable to a governmental entity in the United States, and
access to the US courts for redress of grievances relating to the
performance of its duties.

11. We suggest that the ownership of NEW TLDs be declared public
and explicitly VESTED in the non-profit CORPORATION/IANA.

III. Observations about Specific Comments in the Proposal:
Para. V
A. Stability
The U.S. government should end its role in the Internet number
address systems in a responsible manner..."

These are just words, not substance. You make three assumptions--that
US government should end its role, that it can do so in a responsible
manner, and that your proposal accomplishes that end.
Governance is governance, no matter what entity is carrying out those
functions, private or public. There is no guarantee that the functions
be transferred to private entities will be handled more fairly or with
expense to the public than if the US government were to perform those
functions or contract out the performance of those functions as it did
1993-94 (with certain improvements). Privatization is of interest if it
will work on a fair and competitive basis. However, you are simply
proposing a shift from public to private monopolies in the "registry"
function (which is the focus of our attention, rather than the
function). You will not achieve more stability in this manner. Nor
you achieve a fairer form of governance. We oppose any change which
degrades the quality of governance we should be able to expect from the

Please remember that we are not discussing a function which is difficult
expensive for the US government to perform. It does not require many
people nor much equipment. And, it pays for itself through user fees.
can be contracted out on a set price bid or cost-plus basis at very
expense or run by an entity such as the US Postal Service. This is not
"have to" situation for privatization.

We believe that one or more of the following mechanisms -- utility
regulation pricing, retention of the .com, .org and .net domains by the
or the Corporation or control by a federation of registries will have a
NATURAL COMPETITIVE REGULATION effect upon the multiple new domains
may be run privately -- i.e. they will have to compete against the USG
domains on price, performance and governance, assuming the price of
existing of government TLDs will reflect actual costs rather than a
preferred monopoly position. A mixed system where the USG, IANA or
registrars conferation directly or indirectly manages the existing
might obviate the need for significant governmental regulation of the
registries you propose to create.


Para. V(A)(4)
Seats on the initial board might be allocated as follows:
Three directors from a membership association of regional number
registries, representing three different regions of the world. Today
would mean one each from ARIN, APNIC and RIPE. As additional regional
number registries are added, board members could be designated on a
rotating basis or elected by a membership organization made up of
registries. ARIN, RIPE and APNIC are open membership organizations that
represent entities with large blocks of numbers. They have the greatest
stake in and knowledge of the number address system. They are also
representative internationally."

A word of caution, here. The registries are servants, not masters and
should not be both the governors and the entities created to govern.
should not be represented. Also, IP registries may or may not represent
the interests of the majority of the NET who get their IP addresses
up-stream providers. Further, the supposed justification for IP
representation on the board may not be significant as the rules of
organizations will be rather static and their issues limited. And, they
will have a conflict of interests with the rest of the net when it comes
the issue of expanding the number of registries or encouraging
between registries, which many believe should be done. If they are to
representation, suggest limiting them to one chosen by a different
each term on a rotating basis making their representative (if they
have one) a non-voting member. We also note that these entities are
necessarily "representative," but may have very narrow constituencies.
Those constituencies may be influential in other groups represented on

But, more significantly, we are concerned about where this process is
going. If we are moving toward more significant Internet governance by
this organization in the future, we oppose structuring the board in any
manner other than by representation of specific interests. We oppose
representation of entities which are not chosen by and responsive to
specific interests. Thus, we can imagine (and support the concept of)
representative for ISP organizations from each continent or region of

Two members designated by the Internet Architecture Board (IAB),
international membership board that represents the technical community
the Internet. Two members designated by a membership association (to be
created) representing domain name registries and registrars.

First, we oppose representation for technical groups as opposed to
interests. Second, we would give the domain registries at least as much
representation as the IP registries, and, probably more, since they deal
with broader interests and issues. They should be excluded from voting
upon, if not from discussing, expansion of the gTLDs. Third, we believe
that ISP organizations should have greater representation as they are
predominant consumer and vendor of the various products and services and
protocols involved in the operation of the Internet.

Seven members designated by a membership association (to be
representing Internet users. At least one of those board seats could be
designated for an individual or entity engaged in non-commercial,
not-for-profit use of the Internet, and one for end users. The remaining
seats could be filled by commercial users, including trademark holders.

This is easier said than done as far as providing real representation
the distinct interests identified is concerned. We have come to like
idea of creating a new forum solely for the purpose of electing
representatives, as that resolves the thorny issues of stability and
discrimination between trade groups which might otherwise choose the
representatives. However, we do not know how to define members of the
organization or of the various sections so that one or several interests
are not dominant in voting for representatives of all the interest group
sections. Perhaps "members" would be limited to voting in only one
section. We wonder how you would deal with entities having large numbers
employees who might be members (voters). Thus, implementation of this
section is a concern.


Para. VI(B)
We concede that switching costs and lock-in could produce the
described above. On the other hand, we believe that market mechanisms
well discourage this type of behavior. On balance, we believe that
consumers will benefit from competition among market oriented
and we thus support limited experimentation with competing registries
during the transition to private sector administration of the domain
You recognize but fail to deal with the problems. You simply hope that
will not be manifest during your experimental stage. The problems you
describe are with the system you would create rather than with the
situation which exists. Your proposal is flawed in its conception. You
would create monopolies which you hope to regulate by competition.
"Competitive monopoly" is an oxymoron! You lack the fairy godmother to
force Cinderella's foot into the competitive shoe.

The problem is made intractable because you are counting upon
between unequals. The .com, .net and .org registries already have the
cream of the industry locked up. You only leave stragglers for the
competition. They will never in our lives be able to compete with the
established TLDs on an equal footing in any regard. Suggesting
requires the perspective of an head buried ostrich.

We suggest that the best way to handle this is by some form of
or Corporation/IANA run or licensed registry and shared access to any
qualified registrar for at least the .com, .net and .org domains. Then,
the competitive model has a chance of working for the new domains.
Otherwise, there must be price control/regulation of .com, .org, and


Para. VI(C)

Attempts to impose too much central order risk stifling a medium
like the
Internet that is decentralized by nature and thrives on freedom and

The problem is who should own and profit from particular identifiers and
who should be allowed to get into the registry and registrar business.
Privatizing the function of distributing limited natural resources
(domains) by creating a limited number of private monopoly registries
results in a system which is more difficult to manage and is subject to
greater abuse than retaining control of the registry function by the USG
the Corporation/IANA.


Para. VI(E)
allocation of a portion of the registration fee to the Internet
Intellectual Infrastructure Fund should terminate on April 1, 1998, the
beginning of the ramp down period of the cooperative agreement.

It is time to end the $35 per year fee to NSI and discuss "disgorgement"
the obscene amounts NSI has receive under the $50/yr registration fee
arrangement which was instituted without competitive bid.


1. ...NSI will continue to operate .com, .net and .org but on a
shared-registry basis... The registry will treat all registrars on a
nondiscriminatory basis and will price registry services according to an
agreed upon formula for a period of time.

It is time for the Government to stop buying $750 hammers and giving
gold mines and other natural resources. NSI should not get these
without a competitive bid process. If it is to continue to temporarily
operate the registry as a government contractor managing the public
resource (the domains), the terms of such contract should result in
but not excessive compensation to NSI, and the contract's duration
be reasonably short and periodically reviewed.
Remember, NSI is a new-comer, not the pioneer which developed the
DNS system. There are a lot of others who did more and got a lot less
their efforts. NSI did not create these domains and it was paid
to husband the these resources for the past five years. It has a
and, we assert, a legal duty to return the resource entirely and intact
the government so it may continue operating its domains without a

The Government MUST take immediate steps to deal with what appears to be
emergency created by the end of the "co-operative agreement." It must
start with a demand for copies of the data and programs necessary to
operate the DNS system, arrange for Government or alternative management
begin prior to the end of the agreement, be prepared to seize or
imminent domain to gain control of these assets if it does not get the
unqualified co-operation of NSI, and begin exploring, now, the various
options for managing the resource in a way to maximize the revenue which
can come to the US Treasury from the existing TLDs.


Appendix 1--Recommended Registry and Registrar Requirements
...If, after it begins operations, a registry no longer meets
requirements, IANA may transfer management of the domain names under
registry's gTLD to another organization.

This provision implies (and you must specifically spell it out) that
ownership of the intellectual property associated with the new domains
which are allowed will be vested in the IANA or its successor
The registry should be a contractor or concessionaire, not the owner.


Each top-level domain (TLD) database will be maintained by only
registry and, at least initially, each new registry can host only one

This may not be appropriate if the registry is simply performing the
technical service and control of the TLDs is in some other party.


Appendix 2
4. Alternative Dispute Resolution of Domain Name Conflicts.
a. There must be a readily available and convenient dispute
process that requires no involvement by registrars.
b. Registries/Registrars will abide by the decisions resulting
agreed upon dispute resolution process or by the decision of a court of
competent jurisdiction. If an objection to registration is raised
30 days after registration of the domain name, a brief period of
during the pendency of the dispute will be provided by the registries.

We question the lack of a standard procedure for all US based
We would provide for immediate access to the courts if one of the
disputants feels it appropriate, especially to enjoin a use or
And, there would have to be a tolling of time limits for lack of
of an infringing use or other extenuating circumstances.

IV. Conclusion

The proposal offers a starting point but not a final nor acceptable
resolution of the various issues it addresses. Furthermore, this
was begun to late to deal with these issues in a deliberative and
responsible manner. NSF is responsible for this unfortunate situation
should be brought to task. The only reasonable way to proceed is to
to a cost-plus contractual relationship with NSI while we work through
problems and come up with a broadly acceptable structure to carry us
the future.

We also suggest that the issues of who owns the property bought,
or developed in connection with the co-operative agreement and whether
should be required to disgorge the $100 million it has reaped since
September, 1995, should be referred to the Department of Justice for
investigation and appropriate action.


Internet Texoma,

Eric Weisberg
General Counsel


CC: "Sr. Management" <>