comments Regarding New gTLDs

["Ken Ryan " <ken.ryan@xxxxxxxxxxxxx>]

Paragraph II.C 8 of the Amendment to the Memorandum of Understanding between ICANN 
and the United States Department of Commerce signed in September, 2002 states that 
ICANN should "Continue the process of implementing new top level domains (TLDs)" 
Is there consensus that increasing the number of top level domains is the best or 
only solution to namespace problems on the Internet? Have other relevant ideas 
been evaluated?

In "A Plan for Action Regarding New gTLDs" of 18 October 2002, Mr. Lynne writes 
that ICANN should seek Domain Name Support Organization advice on how to evolve 
the top level generic namespace and summarizes the two primary methods under 
consideration, but also asks "whether to pursue some third alternative." 

I suggest there are advantages in making a slight change to the current addressing 
format and maintaining (or even decreasing) the number of TLDs in use.

By adding a special character and an ordinal to the address string, multiple 
examples of the same name could be registered. As a special character I would 
recommend  # or * which are standard ASCII and recognized internationally since 
they appear on all digital and cellular telephone keypads. The ordinal could be 
numerical, but other notation might also be employed. 

For example: lotus*1.com and lotus*2.com would be separate address strings, as 
distinguishable as smith.com and jones.com. One could address Lotus Cars and the 
other Lotus Software.

The advantages of this format change take longer to present than the concept 
itself. 

  - Personal and company names and trademarks that are actually unique are 
    highly unusual in the real world. The suggested format would allow multiple 
    use of the same company and trademark names while providing individual 
    Internet addressing.

  - The existing TLD namespace would be extended without limit, and without the 
    need to introduce new TLDs. With the exception of personal or family names, 
    the original list of gTLDs presented a comprehensive set of identifiers. 
    Users would not be required to remember new top level domains or asked to 
    accept conceptually indistinguishable TLDs such as "dot com" and "dot biz".

  - Information providers could use their accepted and recognized names as domain 
    names. Current domain name owners would not be required to relinquish their 
    names (and depending on the final technical solution, should not need to 
    append the character/ordinal). 

  - The rational for cybersquatting, cyber-piracy and domain name hoarding would 
    cease to exist. 

  - The original, logical division of domain names by type of information provider 
    would be upheld, while appropriate names in the commercial dot com TLD would 
    be available to all businesses, not just "first come".   

  - Domain name disputes should cease and medium-specific new legislation should 
    be less frequently needed. 

  - The most prevalent Internet domain name software, BIND 9, can handle special 
    characters. Internet users have shown no problem adopting the "commercial at" 
    @ for e-mail and should have no more trouble using # (which already denotes 
    "number" for many people) or * to differentiate between domain addresses. 

  - Users should quickly learn that the quality of information or services presented 
    at a certain address is independent of the ordinal associated with that address, 
    just as television channel 2 is not necessarily better than channel 22,  and 
    2700 Pennsylvania Ave. is a more prestigious residence address in Washington DC 
    than 27, 270 or 2699 Pennsylvania Ave. 

  - The suggested change would provide the name recognition desired by companies 
    for their Internet addresses while defining domain names simply as addresses 
    rather than intellectual property. This was the intent of domain name pioneers 
    such as Dr. Jon Postel. In particular, this format would align the Internet 
    with accepted intellectual property concepts by preventing one registrant from 
    claiming exclusive worldwide rights to generic terms as Internet names.

  - The format supports the registration of many more information, product and service 
    providers in the popular dot com arena, and supports e-commerce competition based 
    on product, price and customer service rather than access to a "good" domain name. 
    Real world legislation regarding fair use vs. protection of famous trademarks 
    could be applied without special regard to the medium.

The suggested domain name format would support "intelligent" information processing 
and piggy-back innovation. For example, if additional information were collected about 
the owners of domain names, their location (for "bricks and mortar" companies) and their 
branch of business or the type of information they provide on the Internet, it would be 
possible to create directories such as "name#0.com" or "name.dir".  In time, these could 
support a search for e.g. dentists in Puerto Rico or identify providers of "Adult 
content".  This function should promote innovation in browser features for managing a 
domain name search or returning to previously visited sites. Extended directories would 
be seen and used by more people than WhoIs listings (which are seldom if ever seen by 
normal users), so discrepancies between directory information and actual content under 
the associated domain name would be easy to catch. Directories of this sort could be 
held to "truth in advertising" standards.  On the assumption that increased 
availability of relevant domain names would lead to more domain name registrations and 
increase the total amount of content accessible through the Internet, this suggestion 
would increase rather than diminish the importance of search engines. User acceptance 
and usability could be easily measured in a test environment, and the response from 
prospective information providers could be collected through e.g. Chambers of Commerce and 
Internet registrars. In vivo evaluation could be carried out on a small ccTLD. 

Based on these considerations, I recommend that ICANN not initiate a new round of 
sponsored TLDs, but rather take a step back and evaluate alternatives to the creation of 
additional TLDs.

Ken Ryan
ken.ryan@operamail.com