Comments on the Action Plan for New gTLDs: Towards an OpenMarket for GTLDs

["Lee McKnight" <LMcKnigh@xxxxxxx>]

Comments on the Action Plan for New gTLDs

December 13, 2002

Dr. Lee W. McKnight
Associate Professor, School of Information Studies, Syracuse University

I am pleased to have the opportunity to contribute my comments to the ICANN Public 
Forum at the ICANN 2002 Annual Meeting, Amsterdam, Netherlands, December 14, 2002, 
with regard to the "New gTLD Action Plan" of 18 October, 2002. I regret not being 
able to submit these by the requested date of December 10, but hope they will 
nonetheless  be taken as constructive suggestions.

By acting in a timely manner on President Lynn’s recommendations to add new 
gTLDs, ICANN will demonstrate its growing ability to satisfy industry and user 
needs through market-sensitive, fair and efficient processes.  My suggestions on 
how the ICANN board and broader Internet community can most effectively rally 
behind President Lynn’s able leadership follow. With the continued growth of 
mobile and nomadic services, it is essential that stable and well-engineered 
namespaces be available from reputable entities. As President Lynn has stated, 
"there is little risk in moving forward with sponsored TLDs to the point of 
full operation."

First, the vision of ICANN President Stuart Lynn in recognizing the need for 
gTLDs is to be praised.

Second, it is also praiseworthy that President Lynn went further, and has 
proposed a plan so that new global Top Level Domains ( gTLDs) can be added 
expeditiously. 

Third, subject to the amendments suggested below being adopted, I believe 
that the gTLD Action Plan of October 18, 2002 could in future be seen as a 
landmark demonstration of the growing maturity of ICANN procedures. Hence, by 
acting expeditiously to adopt an amended Lynn Plan, the ICANN board could 
demonstrate that ICANN has developed well-considered, well-engineered, and 
politically acceptable processes to add new gTLDs. 

I suggest below several specific ways in which President Lynn’s proposal 
could be strengthened still further. If ICANN were to act on my suggestions, 
I suggest that global industry and user needs for new gTLDs would be better 
satisfied.  I focus particularly on Part 1 of the plan; that is, the 
proposal for immediate steps to add new sponsored TLDs; and part III, the 
suggestion that the DNSO's successor choose between a "taxonomic" and an 
"open entry" approach to name space development.  I suggest that a more 
appropriate contrast is between choosing (and therefore having to defend) 
the choice of regulation by random numbers  and a market based approach.  
I personally favor a market based approach, which President Lynn’s plan 
readily accommodates.

Comments on Part 1. Allocating or Marketing gTLDs?

Efforts by ICANN to accommodate demand for access to the DNS name space is 
appropriate. Therefore, I strongly support the recommendation of President 
Lynn that the Board solicit proposals for new sponsored TLDs. 

The only questions then are a) what process could determine what that number 
(of new gTLDs) should be, and b) what number would result from such an open 
public process to determine an appropriate number of new gTLDs that could 
immediately be added with "little risk." The number three is suggested by 
President Lynn, though he states that "it is hard to find anyone who would 
argue that tens or perhaps even hundreds of new small to medium sized gTLDs 
could not be safely added.” as well as his statement later in the report 
that "there was nothing magic in the number seven for new gTLDs. It could 
just as well have been three, fourteen, or twenty."  

Essentially, President Lynn seems to be saying the number three is a random 
number. It is difficult to argue in defense of a process that results in 
random outcomes. It is therefore imperative that the ICANN board assist 
President Lynn in refining the process further.  

However, given the widely recognized need to move some number of gTLDs 
forward as soon as possible, including to meet the needs of worthy applicants 
who were forced to wait due to in retrospect prior random processes, I agree 
with President Lynn that the number of additional gTLDs admitted in 2003-2004 
should be finite.  But rather than predefine the outcome of the review 
process as the number 3, I suggest that it would be better to suggest a 
range of possibilities rather than a specific point outcome. For example, 
to take some upper and lower numbers from the range mentioned by President 
Lynn, I suggest that applicants be told to expect ICANN processes to result 
in between 5 and 15 new gTLDs POSSIBLY being added. Given such a range, 
firms and other organizations may reasonably expect that perhaps 10 dTLDs 
would be established. This would allow as others have suggested, most existing 
legitimate sTLD proposals to be developed, eliminating the need for arbitrary 
decisions to be made. 

What criteria should be applied to those applicants? I defer to the judgments 
which will result from ICANN's increasingly robust and experienced procedures. 
Of course those processes need not prejudge the market for a specific gTLD, 
but need only focus on the capabilities of the proposing entity to provide a 
sufficiently well-engineered service.

Part 3: Towards an Open Market for gTLDs

President Lynn states he will ask the DNSO whether a taxonomic or open approach 
should be used for name space additions. While in other contexts I have found 
taxonomies very useful for analytic purposes, it is rather late in the day to 
suggest that anything other than the market based approach be pursued. In a 
sense the market has always decided - TLDs such as .mil, .edu, and .gov were 
created to meet the needs of different user communities which desired their 
own TLD for one reason or another. Given this legacy, I suspect that a global 
DNS taxonomy that works across languages and cultures to divide the entire 
Internet up is not likely to work. My opinion is that further pursuit of this 
quest would be a distraction for the ICANN community at a time when serious 
efforts must be made to ensure the organization’s future. The only viable path 
forward for ICANN, again in my opinion, is continued focus on ensuring 
well-engineered gTLDs, and not getting distracted by an inevitably elusive 
effort at classification of the Internet.  Leave classification to the (digital) 
librarians!