07-21-97 Electronic Filings on Internet Domain Names

###
Number: 128
From:      Brett M Hogden hogden@rge.com>
To:        NTIADC40.NTIAHQ40(dns)
Date:      7/21/97 9:52am
Subject:   Comments on domain names

-----BEGIN PGP SIGNED MESSAGE-----

[The address to which I am sending was listed in an MSNBC article
http://www.msnbc.com/news/84135.asp, as of 21 JUL 1997.]

I have been passively following the "controversies" for more than a
year, wading through the thousands of e-mail messages sent to the
three or four mailing lists that are most frequently used to discuss
Domain Name system policies, reforms, &c.

I am unsure what form the comments being requested should take.
However, at the very least I would like to submit my comment to the
effect that the gTLD Memorandum of Understanding (MOU) put forth by
the International Ad Hoc Committee (IAHC) seems the most reasonable.
If the required format for comments is something more formal than
this, please let me know.

As a citizen of the United States of America, I am happy to see our
country take an *interest* in how things work out.  I do believe, tho,
that we are just one country in the world and that the Internet is a
world resource.


Thank you,
brett
===
Brett M Hogden, Distributed Systems Architect, Rochester Gas & Electric Corp.
e-mail: hogden@rge.com       vox: +1 716 724 8729        fax: +1 716 724 8227
     Key fingerprint = B1 8D 7E 77 7D 0A 84 AE  C2 49 A4 CC 2E 1D B5 94

-----BEGIN PGP SIGNATURE-----
Version: 2.6.3
Charset: ascii

iQCVAgUBM9NpgpGvcU43X6dhAQFakAQArWYvLeGznYWRq3Ncxk6EzhcsfZsFDnx8
32NHHwkhGcGAXcJksTv0g0L8/nbgen1FeL1fPAIJDDQMLhcpYZeuvyrfAW6tjt97
yt2B6F5OyJAM3JVvSab785pxPk4pFjdwXaAMbJYR/OWyegts7iOC1L/k2WS0WIVy
TlqeLn+PDTg=
=XLyZ
-----END PGP SIGNATURE-----

###
Number: 129
From:      "Joseph Geretz" JGeretz@ucs.net>
To:        NTIADC40.NTIAHQ40(dns)
Date:      7/21/97 12:07pm
Subject:   Net Infrastructure

The best model for the identification of a particular individual or
organization across the globe is already in existence in the form of the
global phone system. Every phone number is uniquely identified by Country
Code, Area Code, Exchange Group and sequence number. This method eliminates
the competitive edge which is arbitrarily assigned by giving a favorable
domain name to a single user across the US while disenfranchising other
organizations with the same business name.

As for searching for a site by user friendly name, each domain should be
allowed the use of up to a certain number of keywords. The same
restrictions on the use of domain names can be imposed on keywords (e.g. no
use of someone else's registered trademark).

I would be happy to discuss this with you in more depth if you wish.

Thank you,

Joseph Geretz
(JGeretz@ucs.net)

###


Number: 130
From:      Carl Oppedahl carl@oppedahl.com>
To:        NTIADC40.NTIAHQ40(dns)
Date:      7/21/97 9:21pm
Subject:   comments

The Department of Commerce has requested comments on the current and
future system(s) for the registration of Internet domain names. 

I am Carl Oppedahl, a partner in the intellectual property law firm of
Oppedahl & Larson.  Our firm's web site is http://www.patents.com>.  I am
submitting these comments via email as requested by the Department of
Commerce.  In addition, these comments are available in HTML on our firm's
web site at http://www.patents.com/nsi/ntia.sht>, with active links to the
URLs cited herein. 

It should be appreciated that this proceeding has the potential on the one
hand to cement into place NSI's monopoly on COM domains and its
multimillion-dollar income stream as well as its flawed policies regarding
trademarks, and on the other hand to remove NSI's monopoly and render its
policies irrelevant.  Thus, the stakes are extremely high.  I have no
monetary interest in the outcome of this multimillion-dollar struggle.  My
interest in this struggle is only as a member of the Internet Community,
and as a critic of NSI's flawed trademark policies.  (An account of NSI's
flawed trademark policies may be seen at
.) I have written two law review articles
that discuss many of these issues:  "Analysis and Suggestions Regarding
NSI Domain Name Trademark Dispute Policy", 7 Fordham Intellectual
Property, Media & Entertainment Law Journal 73 (1996), draft available at
. 

"Remedies in Domain Name Lawsuits:  How is a domain name like a cow?", 15
John Marshall Journal of Computer & Information Law 437 (1997), draft
available at http://www.patents.com/pubs/jmls.sht>. 

As a general matter, I observe that the present comment request from the
Department of Commerce is too little and too late.  This ground has
already been well plowed, and most issues well settled, in the open
comment proceedings of the Internet Ad Hoc Committee
http://www.iahc.org/>.  Thousands of comments were received in that
proceeding and all the issues raised now by the Department were fully
aired previously in the IAHC comment proceedings.  (The comments may be
seen at http://www.iahc.org/contrib/dns-cont.html>.) There is the very
real danger that the Department's present proceedings will harm the
Internet rather than help it, by injecting uncertainty into the process
during the critical remaining months between now and the expiration of the
NSI contract in March of 1998.  It is crucial that the Department conclude
its proceedings and arrive at its recommendations promptly, well before
the end of 1997. 

In the comments that follow, I have set off the NTIA text with ">"
characters so that the reader can see what questions were posed by NTIA
and what answers I have offered.  >A. Appropriate Principles

>Are the following principles appropriate?  Are they complete?  >If not,
how should they be revised?  How might such principles >best be fostered? 

>a. Competition in and expansion of the domain name registration >system
should be encouraged.  Conflicting domains, systems, and >registries
should not be permitted to jeopardize the >interoperation of the Internet,
however.  The addressing scheme >should not prevent any user from
connecting to any other site.  >b. The private sector, with input from
governments, should develop >stable, consensus-based self-governing
mechanisms for domain name >registration and management that adequately
defines >responsibilities and maintains accountability. 

>c. These self-governance mechanisms should recognize the >inherently
global nature of the Internet and be able to evolve >as necessary over
time. 

>d. The overall framework for accommodating competition should be >open,
robust, efficient, and fair. 

>e. The overall policy framework as well as name allocation and
>management mechanisms should promote prompt, fair, and efficient
>resolution of conflicts, including conflicts over proprietary rights. 
>f. A framework should be adopted as quickly as prudent consideration >of
these issues permits. 

These are all important principles.  The Internet has come to be accepted
and embraced by academics, business people, governments, and the world
population generally, in large part because it is *not* controlled by any
one government or post office or corporation or telephone company.  The
early structure of the Internet, designed so that loss of any one node due
to war would not keep the Internet from continuing to function, now serves
admirably to deny any government or post office or corporation or
telephone company a choke-point on the Internet.  This denial of
choke-points has allowed the Internet to serve as a communications channel
for human-rights advocates in troubled countries, and is likely to promote
the free flow of ideas even in countries that resist it. 

The Internet is the strongest technological agent of social change in this
decade, transforming entire industries and media and educational systems. 

March 31, 1998 is a critical day for the Internet.  It is the expiration
day of a five-year contract in which the National Science Foundation
entrusted administration of an important part of the Internet, the
registration of domain names ending in COM, to a contractor called Network
Solutions, Inc.  COM domains existed before NSI entered into its five-year
contract to administer them, and COM domains will exist after the
five-year contract reaches the end of its term.  What will happen to COM
domains on that day?  There are two possibilities:  1.  NSI may continue
administering COM, but instead of answering for its actions to the
National Science Foundation, it will answer to no one.  Such
administration would presumably continue indefinitely.  NSI would be free
to devise new and even more flawed trademark policies, and would be free
to set any prices it chooses for the continued use of COM domains.  NSI
would have a choke-point on the Internet.  NSI, which has always been
secretive, will presumably devise its new policies in secret, disclosing
them to the Internet community only after the policies are set.  2.  The
administration of COM would pass to a competition-based system in which no
single company would have a choke-point on COM domains, a system in which
decisions are made on an open record. 

As I write these comments and consider the two possible futures for the
Internet community, I am reminded that Hong Kong has just a few weeks ago
been absorbed into the People's Republic of China due to the arrival of
the expiration date of a treaty between the United Kingdom and PRC.  PRC
has shut down Hong-Kong's elected government and has replaced it with a
new government, the members of which are appointed by PRC. 

Close study of the text of the five-year contract between NSF and NSI
shows that NSF made clear provision for an orderly succession of power
over COM upon the expiration of the contract.  (The contract may be seen
at http://rs.internic.net/nsf/agreement/agreement.html>.) Under the
contract, NSF can write a letter to NSI, ordering it to hand over "a copy
and documentation of any and all software and data ... in such form and
sufficient detail as to permit replication of the work by a reasonably
knowledgeable party or organization".  Among those members of the Internet
community who have been following this issue closely, this is referred to
as "The Letter".  If NSF sends The Letter, then NSI has to hand over
everything required for NSI's successor to take over the administration of
COM.  If NSF fails to send The Letter, then NSI, due to its unique
position as physical possessor of the software and data, will be capable
of denying anyone else the ability to administer COM, and NSI will
continue indefinitely as the administrator of COM.  NSI will no longer
have a contractual relationship with NSF, and will not be obligated to
answer to NSF. 

Thus, the most critical event of this decade for the Internet community is
whether NSF does or does not send The Letter to NSI.  And if NSF fails to
send The Letter (or fails to do so in timely fashion, which amounts to the
same thing) then that will be the end of the matter.  I was, frankly,
dismayed to learn of this NTIA comment proceeding, and of the recent
establishment of a committee of US government administrative agencies to
discuss the future of the Internet.  I was dismayed because comment
proceedings and government committees have a way of dragging on for months
and then years before anything meaningful happens.  We don't have years,
and we don't have many months, in which to influence the future of the
Internet.  If this comment proceeding is to benefit the Internet community
rather than harm it, this proceeding will have to reach completion well in
advance of the critical date of March 31, 1998.  >B.
General/Organizational Framework Issues

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

At present the Internet has approximately 180 domain name registration
systems, and each has its own advantages and disadvantages.  Most of those
approximately 180 domain name registration systems are associated with
two-letter country codes and are administered within the respective
countries.  As such, they fall outside the purview of this comment
proceeding.  The only domain name registration systems within the scope of
this proceeding are (1) the US domain, and (2) the domains presently
administered by NSI under the contract ending March 31, 1998, namely COM,
EDU, GOV, ORG, and NET.  Nobody fights over EDU or GOV domains;  they are
noncontroversial and the part of the database containing EDU and GOV
doesn't change very often.  What people fight over is COM domains, and to
a much lesser extent they fight over NET and ORG domains.  By far the
majority of Internet domain names are COM domains, which makes it
important to talk about the advantages and disadvantages of the current
domain name registration system for COM.  It is also relevant to discuss
the domain name registration system for US. 

The advantage of the current COM domain name registration system is,
briefly, that the trains are running on time.  People who sign up for a
new domain name often find that the new domain is working within just a
day or two. 

It is also true that DNS lookups are fast and efficient, but this is not
to NSI's credit.  DNS lookups are fast and efficient due to the efforts of
the Root Level Servers, the majority of which are not run by NSI and will
presumably continue to provide fast and efficient service regardless of
what happens on March 31, 1998.  (A list of the Root Level Servers may be
seen at http://www.patents.com/nsi/roots.sht>.)

The chief disadvantages of the current COM domain name registration system
are several.  First, there is the great risk that the present system will
develop into a permanent monopoly held by NSI.  Such a monopoly would
subject the Internet community to NSI's whims regarding trademark policies
and annual fees.  If NSI's track record in these areas were favorable, one
might not fear such a monopoly very much.  But NSI has a poor track record
in both areas.  The chief disadvantage of the current US domain name
registration system is that it forces a geographic specificity that is
unworkable for many businesses.  A company located in Manhattan (and has a
domain name COMPANY.NEWYORK.NY.US) that later moves to Plano, Texas will
not want to change its web sites, email addresses, stationery, and
business cards to COMPANY.PLANO.TX.US. 

Other disadvantages of the current US domain name registration system are
that it is fragmented and that various sub-registrars have unpredictable
pricing for their services.  >2. How might current domain name systems be
improved? 

The main area for improvement would be the transfer of COM away from a
monopoly registrar to a shared-database set of registrars.  A related area
for improvement is the elimination of NSI's flawed domain name trademark
policy. 

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

COM domains should be administered as a shared resource, just as North
American toll-free numbers are shared among long-distance carriers.  The
IAHC proceedings, in which this was all fully considered, recommended this
many months ago. 

>4. Are there decision-making processes that can serve as models >for
deciding on domain name registration systems (e.g., network >numbering
plan, standard-setting processes, spectrum allocation)?  >Are there
private/public sector administered models or regimes >that can be used for
domain name registration (e.g., network numbering >plan, standard setting
processes, or spectrum allocation processes)?  >What is the proper role of
national or international governmental/ >non-governmental organizations,
if any, in national and international >domain name registration systems? 

There was a time when all toll-free numbers in the United States came from
AT&T;  it was a monopoly provider of such services. 

Then the FCC noted that there is simply no good reason for AT&T to be a
monopoly supplier of such services. 

The transition that the FCC followed went like this.  First, the FCC freed
up some numerical combinations (e.g. 800-444 and 800-777) that had
previously been "reserved" and not used by AT&T.  These combinations were
given to competitors such as MCI and Sprint.  During this phase of the
transition, anybody who wished to get a toll-free number from a carrier
other than AT&T was stuck using a number in the range permitted to the
alternative carrier.  For anyone who already had a well-established
toll-free number, such as 1-800-FLOWERS, the "competition" was unhelpful
-- such a customer could obtain non-AT&T service only by giving up the
established telephone number and switching to a different one that no one
had ever heard of before. 

After some years, the transition moved along further, and "800 number
portability" became possible.  A customer that had an 800 number
administered by AT&T could, if desired, switch that number over to MCI or
Sprint.  And it's that way today.  A person hoping to sign up for a new
toll-free number can call any of several carriers, and that carrier (which
need not be AT&T) can check to see if the number is available.  If it is,
then the carrier can place the number into service for the customer. 
Later, if the customer wishes, the customer can move that number to a
different carrier. 

What NSI proposes for the future of COM is that NSI will run COM forever,
answering to no one.  This proposal, if followed, would be very good for
NSI and very bad for the Internet community.  It would freeze the Internet
forever in a terrible condition that is quite comparable to the mid-phase
of the FCC's toll-free transition:  anyone who has a well-established
domain name, such as flowers.com, can obtain non-NSI service only by
giving up the established domain name and switching to a different domain
name that no one has ever heard of before.  This might not be so bad if
NSI's behavior had generally been good.  But, as discussed elsewhere in
these comments, NSI's behavior has been bad.  NSI is forcing a terribly
flawed trademark policy down the throats of its customers.  And NSI
charges annual fees that have been criticized by many.  There is no
assurance that NSI will, in future, confine itself to its present flawed
trademark policy -- for all we know it will continue to revise it and make
it worse.  And there is no assurance that NSI will refrain from increasing
its annual fees, perhaps unreasonably.  The American market for toll-free
telephone service differs from the market for domain name registration
services ("DNRS") in that for telephone service there is a clearly defined
regulator whose power extends indefinitely into the future, namely the
FCC, while for DNRS the only regulator with clearly defined power is the
NSF, and NSF has stated that it will abdicate its power in 1998.  Thus,
while there was "all the time in the world" to set right whatever there
was that needed setting right with telephones, in the case of DNRS we can
take all the time we want as long as we figure it out by about December of
1997. 

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

Many commenters have pointed out that the present fixation by much of the
Internet community on COM domains is, in large part, a result of
unfortunate aspects of the historical management (more clearly,
mismanagement) of the US domain.  If there had been a COM.US domain years
ago, for example, instead of a COM domain, it seems likely that many of
today's difficulties regarding COM would not have come to pass.  COM.US
would be squarely within the power of the US government to control, for
example, where COM is by its nature more multinational (or non-national). 
But history is history;  the past cannot be changed and all we can do is
try to figure out what to do next.  The plain reality is that the Internet
and the Web rely, for their function, on stable domain names and URLs.  It
would be extremely disruptive to eliminate any top-level domain or any
large class of higher-level domains.  The social cost of eliminating
existing COM domains would be profound, in the billions of dollars.  (This
is part of why NSI's monopoly, if permitted to go on forever, would be so
valuable to NSI.) There is much to be said for closing off new COM
registrations and grandfathering the existing COM domains.  This would not
disrupt the Internet or the Web at all, and would have the healthy effect
of diverting future domain name registration activity to the ISO country
code domains.  NSI will, of course, fight this closing-off notion tooth
and nail for obvious reasons.  And the fact is that there probably is a
role in the Internet for non-country-specific domain names, so long as it
can be administered in a fair and reasonable manner.  (For reasons
discussed elsewhere in this Comment, it is suggested that fair and
reasonable administration of COM and other gTLDs is only possible if NSI's
monopoly is kept from continuing forever.) Two futures, then, are
acceptable:  thwarting NSI's monopoly on COM, or closing off new COM
registrations.  The third choice, permitting COM to be monopolized by NSI
forever, is unacceptable. 

>6. Are there any technological solutions to current domain name
>registration issues?  Are there any issues concerning the relationship
>of registrars and gTLDs with root servers? 

The North American Numbering Plan (NANP) for toll-free numbers is a simple
and straightforward example of a technological solution to be emulated. 
There is a central database of toll-free numbers administered by a neutral
party (Database Service Management, Inc., a subsidiary of Bellcore, which
interestingly has recently been purchased by the same company that owns
NSI).  Someone who wants a toll-free number can go to any of several
long-distance carriers (e.g. AT&T, MCI, Sprint) and the carrier makes a
query to the DSMI database and registers the toll-free number.  DSMI's
database is paid for by contributions from the long-distance carriers and
its costs are modest, playing almost a nonexistent role in the prices
charged by the long-distance carriers. 

NSI would have us believe that the database it administers is somehow a
natural monopoly -- that administering it is rocket science and it would
be a mistake ever to change it.  But the database which NSI administers
(under a five-year contract that expires in 1988, recall) is only a little
over a million records.  It would fit easily on the hard disk of a
personal computer.  The NANP toll-free database, by comparison, handles
some ten million toll-free number records.  The database is an order of
magnitude bigger than the NSI-administered database.  It works fine, and
has worked fine for several years now.  It is a "shared" database and the
sharing works fine.  It is evident from this that there is no "natural
monopoly" regarding the administration of COM that should somehow justify
standing idly by and letting NSI have a permanent monopoly on COM.  The
existing technical solutions for sharing North American toll-free numbers
are more than adequate for sharing COM among registrars. 

A related issue raised in this question is the relationship to the root
level servers.  Ever since before NSI's five-year contract began, the
system was set up that the contractor (presently, NSI) would perform its
updates on a single server called the "A" root level server.  Then on an
automated basis, the other root level servers would obtain the updates
from the "A" server.  In this way, any member of the Internet community
would get the same results (e.g. viewing a web site or sending a piece of
email) regardless of which root level server that member might query. 
This system has worked extremely well from the days before NSI to the
present.  A similar relationship exists for NANP toll-free numbers.  DSMI
updates a central server indicating *which* carrier is responsible for
each toll-free number that is in service.  These updates are coordinated
from the various long-distance carriers that offer toll-free telephone
service.  This database is propagated in an automated way to telephone
switches all over North America.  The result is that you can throw a dart
at a map of North America, go to that location, dial a toll-free number,
and get the same result as if the dart had landed elsewhere.  It is
commonplace to order a toll-free number in the morning and have it working
across all of North America by afternoon. 

The shared COM future would work out the same way.  Competing registries,
offering COM domain registration services, would provide inputs to a
neutral registrar.  The registrar would maintain the "A" server (or its
successor if some territory battle with NSI takes place).  The other root
level servers would then take their updates from that server just as they
do now.  Recent service lapses by NSI illustrate that NSI is not
infallible in its administration of the database. 

In early July 1997, NSI made an unpublicized decision to make it
impossible for the general public to download copies of the previously
publicly accessible COM database.  The reprogramming by NSI had the
unintended effect of making it impossible for the root level servers to
obtain *their* updates from NSI's "A" server.  Thus, for several days
members of the Internet community would get inconsistent results when
looking up domain names, depending on which of the several root servers
they happened to query.  People with newly created domain names or
recently updated IP address information were randomly unaccessible to the
public.  It was only after someone outside of NSI (Paul Vixie, operator of
one of the root servers) happened to notice the malfunction that NSI
corrected its programming to permit the root level servers to obtain the
updates.  The incident highlights a striking difference between NSI's way
of doing business and the traditional ways of the Internet community.  The
Internet has heretofore been consensus-driven.  Any change in the software
that supports the Internet has normally taken place only after it has been
discussed by the community.  In this way, imminent mistakes are often
identified and corrected before they cause harm.  In contrast, NSI reached
its decision to change the access permissions to the database in secret,
and did not reveal the change until days later, after Mr. Vixie noted the
out-of-date status of data in the non-A root level servers and asked NSI
what was wrong. 

In another incident, this one in mid-July 1997, NSI corrupted its root
server file.  NSI loaded the corrupted file into its "A" server and passed
the corrupted file to the rest of the root servers.  What followed was a
severe disruption of Internet traffic that reached the front page of the
New York Times.  The disruption was first noticed in Europe, where
early-morning emails and web requests were incapable of being handled.  By
mid-day in the U.S., members of the North American Network Operators Group
(a cooperative body of operators of systems connected to the Internet)
were discussing the problem and trying to figure out how to work around
it.  It appears that NSI first heard of the problem from persons outside
of NSI and corrected the problem only afterwards.  It was not until late
in the day that all of the root servers had been reloaded with current,
non-corrupted data. 

To reiterate, NSI would have everyone believe that administration of the
COM domain name system is difficult and that it would be a mistake to
entrust it to anyone else.  But NSI is fallible, just like anyone else.  A
concern for proper operation of the database should not lead to a decision
(actively, or by inaction) to give NSI a perpetual monopoly over COM.  >7.
How can we ensure the scalability of the domain name system name >and
address spaces as well as ensure that root servers continue to
>interoperate and coordinate? 

Present-day off-the-shelf database tools are scalable to accommodate the
needs of the Internet community.  Those who have programmed the
present-day DNS software of the Internet have the skills and experience to
provide whatever is needed.  Fortunately for the Internet community, none
of this requires NSI's cooperation in case NSI decides to refuse to
cooperate with changes.  >8. How should the transition to any new systems
be accomplished?  Most importantly, all members of the Internet community,
including the US government and its agencies, must support a non-monopoly
plan for Internet administration.  This means reaching consensus early on
that an NSI monopoly is not only unnecessary but must be prevented.  This
means that NSF must write The Letter (the letter directing NSI to hand
over the databases required for a successor registrar to take over COM),
and must do so well in advance of March 1998.  This means that there has
to be a successor registrar.  At present the only viable candidate for
this responsibility is the IAHC/MOU.  Unless some other non-monopoly
entity, acceptable to the Internet community, comes into existence well in
advance of March 1998, this probably means that the IAHC/MOU entity is the
only viable successor to NSI to administer COM, which means NSF must write
The Letter designating IAHC/MOU as the successor.  >9. Are there any other
issues that should be addressed in this area?  >C. Creation of New gTLDs

>10. Are there technical, practical, and/or policy considerations >that
constrain the total number of different gTLDs that can be created?  If the
floodgates were to be opened with creation of a potentially unlimited
number of gTLDs, this would lead to chaos.  To the extent that domain
names do any good, it is because they are capable of being remembered from
the moment one sees a domain name on a television screen to when one types
it in on a keyboard.  At present we have about 180 top-level domains and
it is just barely within human ability to remember a domain name.  Adding
a few dozen more would probably not degrade this substantially.  Adding a
thousand more would make domain names supremely unmemorable. 

>11. Should additional gTLDs be created? 

My view is that there is no problem for which adding gTLDs is the
solution.  The address space of COM is already infinite, and it is just as
easy to add or fiddle with characters to the left of the "dot" as to add
or fiddle with characters to the right of the "dot".  It is also true that
adding a limited number of new gTLDs probably doesn't harm things.  And it
is also true that there are many in the Internet community who feel
strongly that some new gTLDs should be added.  Since it is possible to do
it, and since some people really want to do it, and since it doesn't cause
harm (at least in limited numbers) then we might as well permit it to
happen.  Any new gTLDs must be shared, not monopoly-controlled.  The
IAHC/MOU plan, to its credit, calls for any new gTLDs to be shared, not
monopolized. 

>12. Are there technical, business, and/or policy issues about
>guaranteeing the scalability of the name space associated with
>increasing the number of gTLDs? 

The main issue is that new gTLDs ought not be monopolized.  They ought to
be shared, just as North American toll-free numbers are shared.  At one
time all NANP toll-free numbers began with "800".  Later, "888" was added. 
Wisely, the "888" numbers ended up shared just like the 800 numbers, and
no single long-distance carrier had monopoly control over the new prefix
(call it a gTLD if you wish). 

>13. Are gTLD management issues separable from questions about ISO
>country code domains? 

In some ways, yes, in other ways, no.  A domain name that offends because
it contains a trademark (e.g. EXXON.COM owned by someone other than Exxon)
will not be rendered non-offensive by setting up the domain name to end in
an ISO country code (e.g. EXXON.CO.JP).  All domain names are accessable
from all geographic locations.  To the extent that there is a problem that
needs fixing (in connection with trademarks and domain names), putting an
ISO country code at the end of the domain name does not fix the problem. 
>14. Are there any other issues that should be addressed in this area? 
>D. Policies for Registries

>15. Should a gTLD registrar have exclusive control over a particular
>gTLD?  Are there any technical limitations on using shared registries
>for some or all gTLDs?  Can exclusive and non-exclusive gTLDs coexist? 
There is no natural or system-driven reason why a gTLD registrar such as
NSI should have exclusive control over a particular gTLD such as COM. 
Just as NANP toll-free numbers are non-exclusively controlled by a
plurality of long-distance carriers, so should gTLDs such as COM be
non-exclusively controlled by a plurality of registrars.  >16. Should
there be threshold requirements for domain name registrars, >and what
responsibilities should such registrars have?  Who will >determine these
and how? 

IAHC/MOU and its commenters have devoted substantial time and energy to
these issues.  It would be better not to replow this ground now, but
instead to adopt the recommendations of IAHC/MOU on these issues, or at
least to give great weight to the IAHC/MOU recommendations. 

>17. Are there technical limitations on the possible number of domain
>name registrars? 

No.  Off-the-shelf database products permit databases to be shared by
arbitrarily many entities.  Competent programmers will have little
difficulty scaling the software, and crafting such additional software as
is needed, to handle as many registrars as may come to exist.  >18. Are
there technical, business and/or policy issues about the name >space
raised by increasing the number of domain name registrars?  One issue that
arises (assuming that COM becomes shared) is "what will happen to domain
names that NSI has wrongly placed on hold under its flawed trademark
domain name policy?"  This issue will fix itself, I suspect.  There will
necessarily be a mechanism for domain name owners to transfer the
administration of their domain names from one registrar to another, just
as the owner of an 800 number can switch it from AT&T to Sprint.  The
wronged domain name owners will presumably simply switch their domain
names to registrars other than NSI, who will then reactivate the domain
names. 

This is, of course, little consolation to those domain name owners whose
businesses have been destroyed by NSI's wrongdoing.  But for the ones that
have remained in business despite NSI's wrongdoing, at least it will mean
they can start over in attempting to make use of their domain names. 

>19. Should there be a limit on the number of different gTLDs a given
>registrar can administer?  Does this depend on whether the registrar >has
exclusive or non-exclusive rights to the gTLD? 

There should be no exclusively-controlled gTLDs.  If, somehow, a registrar
comes to control a gTLD exclusively (as NSI does now and proposes to do
forever with COM), then that registrar should not be allowed to control
any additional TLDs. 

>20. Are there any other issues that should be addressed in this area? 
>E. Trademark Issues

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

Internet conduct is subject to trademark laws, just like any other
conduct.  Someone whose conduct violates trademark rights will answer for
it in court, whether that conduct has to do with a toothpaste package, the
text of a web site, a fourth-level domain name (e.g. exxon.oil.com.us), or
a second-level domain name.  It is aberrational that so much attention has
fixated on second-level domains when there are so many other ways for
someone to infringe a trademark via Internet conduct.  The main reason
that so much trademark attention has fixated on second-level domains is
NSI's flawed policy, first put into place in July 1995.  NSI's policy has
wrongly prompted trademark owners everywhere to believe (or to pretend to
believe) that mere text identicality between a trademark and a domain name
gives rise to the power to cut off and take away the domain name.  While
the courts have slowly tried to get out from under the wrongly-directed
momentum created by NSI (most recently in the Lockheed case cited below)
the harms caused by NSI's policy will take a long time to be undone.  >22.
Should some process of preliminary review of an application for
>registration of a domain name be required, before allocation, to
>determine if it conflicts with a trademark, a trade name, a geographic
>indication, etc.?  If so, what standards should be used?  Who should
>conduct the preliminary review? 

The Internet, for better or worse, is non-geographic in nature.  Any web
site anywhere in the world can arouse the anger of any trademark owner
anywhere in the world.  Nothing about a domain name (even if it is an ISO
domain name) can insulate a web site owner from legal liability if the
conduct gives rise to liability that a court is willing to impose and
enforce.  Thus, a domain name registrar that might set a goal of screening
all domain name applications for trademark trouble would face the task of
searching the trademark records of all of the 180 or so countries in which
there are trademark systems.  The majority of those countries do not have
online records and the only way to do trademark searches is by going in
person and physically searching physical (paper) records. 

In the US, at least, the law is clear that such searching is unnecessary,
and that a domain name registrar is under no general duty to screen
applications for trademark problems, any more than a stock exchange would
be under a duty to screen proposed stock trading ticker symbols for
trademark problems.  See Panavision v. Toeppen, 96 cv. 3284, US. Dist.
C.D. Cal. December 2, 1996, at 10 "NSI is under no general duty to
investigate whether a given registration is improper [citing MDT]".  See
MDT Corp. v. New York Stock Exchange, Inc., 858 F. Supp. 1028, 1032 (C.D.
Cal. 1994) which held that the New York Stock Exchange was under no
general duty to screen proposed stock exchange ticker symbols for possible
trademark infringement.  In the US the law is also clear that mere text
identicality between a domain name and a trademark does not give rise to
liability under the trademark laws: 

     The Court begins with the observation that "unlike 
     a patent or copyright, a trademark does not confer 
     on its owner any rights in gross or at large." MDT 
     Corp. v. New York Stock Exchange, Inc., 858 F. 
     Supp. 1028, 1032 (C.D. Cal. 1994) (citing Traeger 
     v. Gordon-Allen, Ltd., 71 F.2d 786, 768 (9th Cir. 
     1934)); see also Anheuser-Busch, Inc. v. Balducci 
     Publications, 28 F.3d 769, 777 (8th Cir. 1994) 
     (noting that "unlike copyright and patent owners, 
     trademark owners have no rights in gross"). Therefore, the law does
not per se prohibit the use of trademarks or service marks as domain
names. Rather, the law
     prohibits only uses that infringe or dilute a trademark or service
mark owner's mark. Moreover, innocent
     third party users of a trademark or service mark have no duty to
police the mark for the benefit of the mark's owner. MDT, 858 F. Supp. at
1034. Consequently,
     the mere fact that a person registered a SKUNK 
     WORKS or a variation thereof as a domain name 
     does not mean that the person infringed or diluted 
     Lockheed's mark. (Lockheed Martin Corp. v. NSI, 
     96-cv-7438, C.D. Cal., March 19, 1997.)

It is ridiculous to suggest that a domain name registrar should screen
domain name applications, one by one, for possible infringement.  As the
Lockheed court indicates, a text match does not mean that there is
infringement or dilution. 

> If a conflict is found, what should >be done, e.g., domain name
applicant and/or trademark owner notified >of the conflict?  Automatic
referral to dispute settlement?  From the preceding discussion, it's clear
that the domain name registrar would not and should not be "finding"
conflicts.  It is up to a trademark owner to police the world of domain
names just as with toothpaste packages or any other possible trademark
use.  If the trademark owner purports to find fault with a particular
domain name, the registrar should play no active role other than providing
contact information between the parties (see RFC 1591).  This has been the
rule on the Internet since before NSI crafted its flawed policy. 

It's also important to distinguish between "conflicts" that are perceived
by a trademark owner at the time a domain name is registered, and
"conflicts" that are perceived months or years later.  If a domain name
owner has been using a domain name for many years, it is unfair to propose
cutting off the domain name simply because someone who sat idly by for
those years now suddenly wishes to assert that there is a "conflict". 

A domain name registrar that receives a brusque letter from a trademark
challenger demanding that the registrar cut off a domain name, should
politely but firmly decline to sit in judgment on the domain name but
should instead invite the would-be challenger to present its claim in a
competent tribunal, such as a normal court or, perhaps, a WIPO tribunal. 
>23. Aside from a preliminary review process, how should trademark rights
>be protected on the Internet vis-a-vis domain names?  What entity(ies),
>if any, should resolve disputes?  Are national courts the only
>appropriate forum for such disputes?  Specifically, is there a role for
>national/international governmental/nongovernmental organizations?  The
one thing that may be said with confidence is that NSI is the wrong party
to decide such disputes.  NSI has a long track record of frequently
getting the wrong answer as to what should be done when a trademark
challenge arises. 

As for the selection of forum, one must not forget that if a dispute has
to do with, say, a toothpaste package, the only generally applicable forum
is a national court.  The same is true on the Internet for all types of
disputes relating to web site content, third- or fourth-level domain
names, email content, news posting content, and almost everything else
that is capable of being done on the Internet. 

It is odd that of all the things that can be done on the Internet, exactly
one (that is, the selection of a second-level domain name) is thought by
some (including NSI) to be peculiarly appropriate for extrajudicial
treatment. 

The IAHC and MOU proceedings have proposed a WIPO-based dispute resolution
mechanism.  As with anything new, it is not easy to know or to predict
with certainty how that mechanism will work out in practice.  There is
really no way to find out other than to try it and see if it turns out to
be fair and just.  It would surely be fairer and more just than what NSI
does now.  >24. How can conflicts over trademarks best be prevented?  What
>information resources (e.g. databases of registered domain names,
>registered trademarks, trade names) could help reduce potential
>conflicts?  If there should be a database(s), who should create >the
database(s)?  How should such a database(s) be used?  The best
preventative step regarding conflicts over domain names is education of
the legal community and of the Internet community.  Lawyers and lay people
need to read and re-read the quotation above from the Lockheed case.  It
is a simple fact of life that the majority of trademarks are non-unique,
that is, more than one person or entity uses the mark.  All but one of
them will have to survive without ownership of "mark.com".  Text
identicality does not mean trademark infringement or dilution, and someone
who has been using a domain name for years without infringing anybody
else's trademark rights should not have to give it up simply because a
latecomer to the Internet covets it. 

Although the vast majority of trademarks are non-unique, a very small
fraction (well under one percent) are unique.  Unique and coined marks
such as Exxon and Kodak and Xerox may be created and asserted without
depriving anyone of the use of dictionary words, and such marks probably
deserve strong protection generally.  See particularly my article
"Remedies in Domain Name Lawsuits:  How is a domain name like a cow?", 15
John Marshall Journal of Computer & Information Law 437 (1997), draft
available at .  It is also
well-established that the databases of domain name registrations need to
be public records.  There is a great need for people to be able to search
the databases for many reasons including checking for possible trademark
infringements.  NSI has urged, rather unconvincingly, that it "owns" the
COM domain name database.  As a consequence of this assertion of
ownership, NSI has refused to make the COM database available to trademark
search firms and has intimated that this ownership interest somehow means
that no other entity can ever administer COM without NSI's permission
(which permission NSI has said it will not give). 

The ownership claim by NSI fails for many reasons.  First, NSI is merely a
five-year government contractor.  The database existed before NSI began
the contract, and will exist after the contract runs out.  Under the
contract, NSF (as discussed above) can simply order NSI to hand over the
database to the successor registrar. 

Second, the database was authored not by NSI but by the hundreds of
thousands of domain name owners who authored the records that make up the
database.  NSI's contribution was little more than alphabetization of the
records.  Under the Feist US Supreme Court case, the copyright law does
not give protectable rights to NSI. 

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

Some registries do this now.  See the comprehensive survey of registration
policies compiled by Geoffrey Gussis at
.  >26. How would the number of
different gTLDs and the number of >registrars affect the number and cost
of resolving trademark disputes?  The effect of adding TLDs or registrars
would be small compared with the effect of educating the public and the
legal profession about trademark law, and the effect of rendering NSI's
flawed policy irrelevant by keeping NSI from getting a permanent monopoly
on COM.  It might be thought that "cybersquatters," that is, people who
register domain names identical to unique or coined names, will see new
gTLDs as a fertile new ground for cybersquatting.  There are two reasons
this won't happen.  First, what must not be forgotten is that there are
already over 180 TLDs, and cybersquatters tend to prowl only in COM. 
Adding a few new TLDs won't change the fixation of many on COM to the
exclusion of other TLDs.  Second, the window of opportunity for
cybersquatters has come and gone.  COM became trendy several years ago,
and some cybersquatters picked up on it before some big companies with
unique or coined names.  But that won't happen again -- the "trendiness
gap" between those who perceived the value of the Internet and those who
didn't, has closed.  Any company with a coined or unique name that feels
it must have a domain name (that is, a COM domain name) has already done
so.  >27. Where there are valid, but conflicting trademark rights for a
>single domain name, are there any technological solutions?  The question
seems poorly stated.  Valid trademark rights don't conflict; this is part
of the definition of valid trademark rights.  Perhaps the question can be
rephrased as "when there are two or more parties that wish they could have
the same domain name, what can be done?"  One of the difficulties about
this question is that in general, one of the parties registered the domain
name and only later did the other party come to wish to possess the domain
name.  I have been involved in a number of settlement negotiations between
such parties, and in several cases the parties arrived at comfortable
resolutions such as agreements to provide cross-links each to the web site
of the other. 

Some commenters propose that neither party, in such a case, be permitted
to use the domain name, and that some sort of "pick the company you want"
menu be provided in place of the web site previously provided by the (now
ex-) domain name owner.  To force such a change would be wrong.  It
disserves the Internet community because it breaks previously functional
links, and disrupts the flow of email.  It ignores the valuable goodwill
that may have accumulated around the domain name, web address, and email
addresses. 

The public, and the legal profession, are slowly coming to understand that
search engines and directory services exist, and that the day has passed
when a company with a non-unique, non-coined name could hope that people
would guess at the domain name by typing the company name into the browser
window.  Search engines, voluntary cross-links, and the like will reduce
the need for "guessability". 

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


###

07-24-97