Re: IP/TM Concerns & New GTLDs (was:Re: [wg-c] Who should vote for new gTLDs)

["Mark C. Langston" <skritch@home.com>]

On 29 July 1999, "Kevin J. Connolly" <CONNOLLK@rspab.com> wrote:

>
>The trademark infringements that the TM/IP community is concerned 
>with are not in the root, they are in the Second Level (and Higher) 
>domain names.  There is already a tremendous transactional cost in 
>searching the big three and the significant ccTLDs.  Adding 100 gTLDs 
>(which are perceived as having more acceptability than most ccTLDs) 
>will exacerbate the problem.

Sorry, I misspoke (errr...mistyped).  Where I said roots, I meant
to say registries.

I don't see how gTLDs "have more acceptability" than ccTLDs, except
that the average net user has no idea that the ccTLDs exist, except
for some of the more prevalent ones (.fr, .il, etc.).  Even then, I'd
bet that many think .il is Illinois. :)  

My point here is that much of the data the TM/IP interests want to
have easy, quick access to is personal contact information.  Now,
are we working under the assumption that SLDs are property?  If so,
are we working under the assumption that they're real estate?  I don't
see why anyone has the right to my personal information for
a private transaction, unless it's for real property, in which case
it's supposedly a matter of public record.

Also keep in mind that there's still a significant movement to prohibit
unsolicited commercial/bulk e-mail.  Some of these TM/IP interests may
simply farm the registry database for UCE purposes, or even for physical
junk mail.  There are privacy issues to concern ourselves with, and on
that I will be absolutely inflexible.  If the TM/IP interests can't come
up with a way to protect their interests while respecting personal
privacy, I am unsympathetic.

>
>> They're all more or less
>>searchable.
>
>Most of them are less searchable, if only because there's no uniform 
>protocol.  And there's no assurance that future gTLDs will be as 
>searchable as .com <smirk>
>

There's no assurance they'll be less searchable, either.  Once again,
matters such as the ease with which someone can grep a registry's DB
are privacy concerns.

Hm.

I actually just had an idea (Yes, it's frightened and confused, and
in an entirely foreign territory).

How about something like this:

1)  TM/IP interests submit to the registries a list of strings that
would, if used, directly infringe upon their TM/IP concerns.

2)  Whenever a registry's database is updated, the new additions are
compared against these strings.  It's easy, computationally inexpensive,
and exactly what computers are good at.

3)  The potentially conflicting SLD is flagged by the registry, an 
e-mail is sent by the registry to the registrant and the TM/IP interest,
and (assuming there's a mandatory DRP in place) potentially to
the arbiter.

4)  The TM/IP can then decide if it actually is infringement and act
accordingly, and the registrant is notified that they've registered an
SLD that may be contested.  Heck, you could even give the registrant
the option of resubmitting the application with a new SLD, to avoid
possible infringement and arbitration.

This strikes me as fair on all sides.  The TM/IP interests don't have
to waste time and money searching all the various registries.  The
privacy advocates don't have to worry about the TM/IP interests having
full access to the registies' contact database.  And the registrant
is notified ahead of time that their purchase of the SLD may be questioned,
and would be allowed to act accordingly.  The registrar sets up the
registry to compare a list of strings against the diff of the database,
which is not that costly, computationally or otherwise, and send out
e-mail if a conflict is found.

Would everyone be willing to accept a greater number of gTLDs if something
like that were in place?  


>As I have said before, the ADR process is a necessary but not sufficient 
>condition for approval by the TM/IP community of new gTLDs.  The 
>increased transactional cost brought about by pell-mell enlargements of the 
>TLD name space still threatens to overwhelm the process.

I think my suggestion above would go a ways towards reducing this.
Many people (the honest ones, anyway) would probably back off if they
were told that the SLD they just tried to register is stepping on
$BIGCOMPANY's toes.  That just leaves the squatters, who probably don't
have the cash to fight it out.

Many of the cases I've seen/heard about aren't instances of willing
disregard of TM or IP.  Usually, someone registers an SLD, establishes
a presence, and then weeks, months, or even years later, a TM/IP 
interest speaks up.  At this point, the holder of the SLD, who didn't
mean to infringe, has invested quite a bit in their presence under that
domain.

If they were told up front that it would be in question, I'd bet
that many of these cases would never have happened.

Now, there will be cases where the infringement is questionable at
best; in these instances, don't be surprised if a few people decide they
want it arbitrated or litigated.  But then, that's what the processes
are there for.  

>
>Again with the root servers?  Nobody takes the prospect of "infringing" 
>gTLD names seriously.  The infringements will come about when SLD 
>names are registered in the new gTLDs.

Sorry.  Again, I meant to say "registry" where I was saying "root".

>>
>>That leaves the expense.  To a certain extent I believe that the
>>expense should be accepted as part of the cost of maintaining 
>>exclusive use.  If the benefits of exclusive use are outweighed
>>by the cost of maintaining it, perhaps nonexclusivity should be
>>considered.
>
>This would make trademark the exclusive purlieu of the very rich 
>and powerful.  This would not be good for consumers, the world 
>economy, or the growth of e-commerce.  The vast majority of trademarks
>are held by little guys trying desperately to establish brand recognition
>in a marketplace filled with big marketing powers.

True.  I sometimes forget that Mom 'n' Pop, Inc. also hold TMs.
However, let me ask you this:  Is the cost of litigation less than
the cost of registering a domain?  It seems reasonable to assume it's
much, much more.  If we put into action something like I outlined
above, not only would the TM/IP interests be notified before the
domain is registered, they'd then know there's interest in such a
domain, and perhaps they'd just pay to register it themselves instead
of paying to fight over it every time someone tries to obtain it.


[...snip more confusion created by me not saying "registry"]

>
>>Additionally, I'm entirely unclear as to how you propose to
>>ascertain intent from such a database.

> There will of course be more than one database.  Some registries
>will permit o r require a statement of purpose when registering a SLD
>name.  Other indicia of intent include the content of the website (if
>any) or the absence of a website associated with the domain name.

That's up to the registry.  However, I'll bet you a shiny new donut 
(with sprinkles, even) that those registries don't become very popular.
But then, that's why it's up to the registries.  It's a business 
decision.

However, would you expand on how "the absence of a website associated
with the domain name" confers intent, in any way?  The only intent I
read from it is that the registrant either doesn't have a web site
for the domain, or doesn't want one.  And here's where I'll repeat
my daily mantra:  the namespace is more than just a place to hang
a webpage.  There are literally dozens of legitimate uses for a
domain name that have nothing at all to do with web pages.


>>Would the TM/IP community be agreeable to a plan which set an upper
>>limit by a certain date, with a slow timetable for expansion?
>>E.g., 2 or 3 new TLDs a month for several years?  One every
>>two months for 5 years?  Something along those lines?
>
>You'd have to ask them.  I'm not a member of the TM/IP constituency.

Yes, but you seem to be speaking for them, by fiat.  You're arguing
their case, and they're not speaking up, even though the concerned
constituencies have representatives as members of this WG.  So feel
free to answer with the caveat that you're not a member of the TM/IP
constituency.

>
>>
>>There's still the burning issue:  which TLDs?
>
>It's pretty much a foregone conclusion that this WG will not even attempt to
>propose a list of gTLDs or an implementation schedule within the time frame
>set by the Names Council.  So we would probably do better to consider the
>broader policy issues surrounding gTLDs.

Granted.  But then the question becomes, who decides?  If this WG can't
do it in the time allocated, there really should be another WG to deal
with this, with a later deadline.

-- 
Mark C. Langston	     			Let your voice be heard:
mark@bitshift.org				     http://www.idno.org
Systems Admin					    http://www.icann.org
San Jose, CA					     http://www.dnso.org