Re: [wg-b] me thinks thou dost protect too much ...

["Mark C. Langston" <mark@bitshift.org>]

On Mon, Apr 10, 2000 at 10:02:16AM -0400, Hartman, Steve wrote:
> I diagree with your statement that ownership of domains does not bear on the
> Internet's integrity. To give one example, Nabisco received numerous
> complaints at one time from consumers whose children, presumably looking for
> the producer of Oreo cookies or the information about them, typed in
> <oreos.com> only to be pointed to a pornographic site. The usefulness and
> value of the Internet as a communicative tool is decreased by that use. The
> issue, to my mind, is whether the costs of allowing certain domain names to
> be monopolized in some TLDs are justified by the benefits.
> 


The usefulness and value of the Internet as a communicative tool will
be decreased by the IP interests superimposing their political and
financial will onto a functioning technical system.

Consider:  Right now, we have:

 namespace -- a term describing a set of identifiers whose sole
purpose was to provide a ***hierarchical*** indexing of dotted-quad
addresses.  In namespace, for the most part (but not always) there was
to be a one-to-one, or at worst, a one-to-many or many-to-one mapping
of namespace identifiers to dotted-quad addresses.

 TMspace -- a term describing a set of identifiers that embody a large
many-to-many mapping of labels to entities.

...and now, because the registrars aren't marketing domain names as
unique within a TLD, but are in effect making the SLD the top node of
the tree of unique identifiers within a subset of names, the IP interests
want to force that many-to-many mapping onto a system that de facto is
not organizaed that way, and to hell with the consequences of doing so.

The fact that some people prefer to try to guess at domain names by
assuming every entity on the planet has a domain name in the form of
"canonicalproductname.com" is ignorance and naivete on the part of the
user.  To pander to that ignorance by trying to map two incongrous sets
onto one another is just compounding that ignorance.

Or, from another angle:

There are a limited number of words in the English, and in other,
languages (I will use English in my examples because it is both my
native language and the Lingua Franca of the Internet at the moment).
The governmental offices that deal with IP interests worked around
this problem by developing product category boundaries and
geographical boundaries which limit the reach of trade and other
marks.  While I cannot presume to speak for how an IP interest must
feel about these limits to the reach of their mark, I can say as an
outside observer that the system works, and works well.  [NOTE: By
'outside observer' in the previous sentence, feel free to read,
'consumer', because that's exactly what I am.]

Now, we have namespace, which does not on the surface appear to have
these limits and boundaries.  But in reality, they do: The TLDs
themselves are natural boundaries and serve to uniquely identify those
SLDs under it.  However, a certain registrar has chosen to blur that
distinction in its marketing, and the flood of new registrars that
ICANN has allowed to be created have chosen to carry on that
tradition.  So, now that we are faced with the impending introduction
of new gTLDs, everyone's panicking and assuming that these gTLDs will
be marketed in this manner as well.  So, instead of attacking the true
problem -- that the registrars have deigned to market the TLDs in such
a way that the strings are completely irrelevant, and thus confused
the public into thinking that everything is ".com" -- you'd rather use
significant legal muscle and political clout to impose what amounts to
a mathematically impossible mapping of the space of existing marks
onto the space of domain names, acting as though the top-level unique
identifier is the SLD rather than the TLD.

This is how I perceive the entire issue, and it doesn't sit well with
me.  It is, to me, a poor solution to the wrong problem.  The IP
interests could have had this problem solved by now, at considerably
less effort and expense, if you'd chosen instead to work with the
registrars and help them see the benefit in marketing TLDs as unique,
rather than as 'Yet Another .com.'

Of course, now the registrars will chime in and complain that this would
go against their profit model.  Somehow, the ccTLDs seem to manage
without trying to be the next catch-all for SLDs that already exist in
.com.  Some, in fact, are trading on the uniqueness of their TLD.
While I don't necessarily approve of the particular instance, the .tv
ccTLD is a good example of what I mean when I refer to 'trading on
TLD uniqueness'.  Even at their exorbitant prices, I saw a report this
morning stating that the registry for .tv had sold $300,000 worth of
domain names.



> Cybersquatting of well known names fosters miscommunication and is
> exploitative.Who should be entitled to reap the benefits (whatever they may
> be) of names such as <nabisco.com> or <billgates.com>? Those domains have
> recognition value because of the efforts of Nabisco and Bill Gates. If
> others believe they are entitled to own them, they ought to bear the burden
> of demonstrating a right or consumer benefit to use those domain names
> greater than Nabisco's or Bill Gate's and the confusion such unauthorized
> use causes.
> 
> Steve Hartman
> Nabisco, Inc. 

Steve, it's not my contention that IP interests shouldn't be allowed
to have _any_ registration.  Indeed, I'll support Nabisco's efforts to
own oreos.com.  It's the heavy-handed, across-the-board, "we're going
to prevent ownership of the string 'oreos' in all 240+ TLDs" approach
that prevents anyone, for any reason whatsoever, from every owning
oreos.ANYTHING that I take issue with.  It goes against common sense,
it runs contrary to reason, and it ignores years of thought in TMspace
as regards the limits of scope for marks.

There's a place for everyone in namespace, believe it or not.  There's
a reason behind every name registration, and it's not always bad faith.
Right now, the tide of opinion in this working group and in ICANN in
general seems to be that ANY registration that has substring overlap
or a significantly small linquistic distance to an existing mark is
'cybersquatting'.  Even in the WIPO arbiter's reports, assumptions
are being made about the content of websites, the _existence_ or 
_nonexistance_ of websites or other services within a domain, the
corporate status of the owner, and so on as indicators of bad faith
registration.  

If things continue in this manner, there will one day be no room for
anyone but corporations.  While this may not seem like a bad scenario
to you, it's devastating to the Internet's usefulness and value as a
communicative tool.  Two things to keep in mind:

1)  Domain names are not, should not be, and were never meant to be,
    a communicative tool qua domain names.  It is only that marketing
    has made it so.  Those of you who sincerely believe that marketing
    departments have any but the most tenuous grasp on reality, please
    raise your hands.

2)  It's all about the content.  And it's all about the use.  And the
    content can be communicated by means other than web pages.  But
    it's beyond any arbiter's legal power to accurately determine the use
    to which the IP address(es) answering under a particular SLD have
    been put, and confiscation of domain names in addition to the alteration
    or elimination of content when the name is not an exact match for
    the mark in question is confiscation of property, and in my opinion
    a significant overstep of authority.


-- 
Mark C. Langston
mark@bitshift.org
Systems & Network Admin
San Jose, CA