Re: News update

[Robert Shaw <robert.shaw@itu.int>]

Phillip C. Reed wrote:
> 
> I have yet to see any evidence that would indicate that more TLDs, hence
> more domain names, would make things any better.

The logic goes like this:

1. TLDs combined with SLDs are unique identifiers

2. But the non-DNS world is based on non-unique names. For example, the 
   multi-dimensional space of (largely national) trademark law is a 
   reflection of the reality of a shared real-world name space (e.g., x 
   numbers of registrations of the word "Genesis" in one of the 42 Nice 
   trademark class). Needless to say, the intersection of national trademark 
   law with gTLDs make this problem worse.

3. Reflecting this shared real-world name space requires additional identifiers
   to uniquely differentiate entities in the DNS.

5. This can be accomplished by introducing additional qualifier or hierarchy 
   entry points, namely:

   a. introducing subdomain granularity (e.g., delta.air.com, delta.30.com)
   b. additional textual qualifiers in an SLD (e.g., delta-air.com)
   c. numerical identifiers in an SLD (e.g., delta4.com)
   d. numerical identifiers in the TLD (e.g., delta.c41)
   e. additional gTLDs (delta.air)

6. Because of:

   a. Market preference for short and mnemonic names
   b. Installed base of names using the current gTLD naming convention
   c. History of the gTLD name space with no subdomain granularity
   d. Unlikely market acceptance of numerical qualifiers/hierarchies 
      (e.g., delta44.com)
   e. The importance that entities attach to using their 'unique' names (in 
      their world/context)
   e. Unlikely market acceptance of non-meaningful gTLDs (e.g., delta.c41 even
      if this might be the most scalable)
   f. proof of a-e above through many conflicts over rights to names even when
      global Internet penetration is still statistically small.
      
Leaves 5(e) as the only acceptable solution.

However, this causes creates pain for trademark holders who daily have to deal 
with entities registering their trademarks in different venues either out of 
malevolence or ignorance. The international legal trademark framework is 
quasi-completely unprepared to deal with this phenomenon (e.g., there are no 
substantive guidelines as to what constitutes a strong or famous trademark at 
the international level). This had made it extremely difficult to hammer down 
the Administrative Domain Name Challenge Panel (ACP) substantive guidelines
(see http://www.gtld-mou.org/docs/dispute.html and specifically
http://www.gtld-mou.org/docs/notice-98-01.html).

As new gTLDs are introduced that over time that necessarily become more specific 
in character (e.g., .info vs .sports), this provides a form of quasi-trademark 
name space mapping (e.g., if prince.sports had been available, would Prince
US have attempted to litigate over prince.com?). It's clear that the dispute 
resolution framework will have to evolve next to accomodate the granularity 
introduced by more specific gTLD names.

Robert
-- 
Robert Shaw <robert.shaw@itu.int>
Advisor, Global Information Infrastructure
International Telecommunication Union <http://www.itu.int>
Place des Nations, 1211 Geneva, Switzerland