Minds + Machines Comments on the new Draft Application Guidebook

[Antony Van Couvering <avc@xxxxxxxxxxxxxxxxxxxx>]

Thank you for the opportunity to comment.

First, we would like to thank ICANN staff for their diligent work in  
producing a much-improved draft of the Application Guidebook.   While  
the new guidebook is much improved, we believe there is room for  
additional improvement in the following areas.

I.  Evaluation Panel Procedures:
At present, there is no published procedures on selecting Comparative  
Evaluation panelists.  In order to preserve fairness, we would like to  
see clarification of how panelists will be selected.  Specifically, we  
would like to see ICANN conduct a conflict of interest check.  Given  
the small size of the ICANN community (and therefore, of qualified  
panelists), the risk of conflict interest is great.  We would like  
ICANN to conduct (and stand by) a conflict of interest check, and  
require panelists to provide a written conflict of interest statement  
specifying where they might have a conflict.
II.  Scoring for Community Applications
A score of 12 (instead of 14) should be sufficient to be qualified as  
a community under the Comparative Evaluation procedures.  We are  
concerned that the current high bar, although useful for making sure  
that the TLD serves the community, is so restrictive that it may cause  
registries to fail because they will be unable to attract qualified  
applicants, for two reasons:  (1) the vetting of applications would be  
so cumbersome that it would discourage registrations, and (2) that  
qualified applicants would be refused.  For instance, if .NYC wanted  
to score a 14 by restricting registrations to registered residents, it  
might be unable to allow native-born New Yorkers who now live in  
Florida to register under .NYC, or to allow companies headquartered  
elsewhere to register, even though they have significant presence in  
New York City.
III. String Contention
We are concerned that current string contention policies based on  
meaning (as opposed to string similarity) may discourage valid  
applications.  The best way to illustrate this is through an  
example: .VIN vs. .WINE. These two strings, while referring to the  
same commodity, could serve entirely different needs and communities.
There is a currently announced .VIN, which would serve the French wine  
community.  A .WINE might serve the U.S. wine community.  There is no  
necessary conflict between these two strings, which serve different  
communities.
Therefore, if two or more applications serve, and are backed by, two  
different communities, ICANN should find a way to allow both.
IV. Auction Procedures
We are concerned about gaming of the auction procedures, and about the  
possibility of illegitimate auction contenders.  Let us suppose that  
an applicant for .STRING realizes that there is going to be an  
auction, and that while they do not have the funds to win, they wish  
to (a) drive up the cost for the competition, or (b) get paid off to  
go away.
To prevent this kind of behavior, we propose the following:

1. ICANN should require proof of ability to pay
2. 20% of each bid increment should be non-refundable to prevent non- 
serious bidding.
3. The parties should be contractually obligated to pay in case they  
win.
At present, there is no disincentive to drive up the bidding by a non- 
serious party.
V. Existing ccTLDs and Geographic TLDs
There may be legitimate applications for territory, country, or  
regional gTLDs which may appear to conflict with existing ccTLDs.  But  
because existing ccTLDs may have very high prices, or very restrictive  
registration policies, the relevant governmental authority, along with  
the community affected, may strongly support a new gTLD.   If a  
community and/or government authority wishes to support a gTLD  
application, this should be allowed.
VI. Multiple labels for the same geographical term
From both an IDN and ASCII perspective, current ICANN rules require  
double (or triple) payment.  For instance, .koln, .köln, koeln,  
cologne.  Cities or regions which are widely referred to by multiple  
labels should be given an opportunity to use all of these labels for  
the same price.
VII. Rights Protection Mechanisms
We are highly encouraged by the procedures recommended by different  
parties to help solve the besetting problem of protecting the rights  
of intellectual property owners.  First, the trademark validation  
database put forward by Bart Lieben of Lada is, we believe, a way to  
provide for  inexpensive and error-free Sunrise periods (or similar  
pre-launch registration opportunities).  Second, Demand Media's rapid  
takedown proposal offers a less-expensive and far quicker method to  
deal with obvious cybersquatting attempts.
With thanks again for the Guidebook and the process for improving it,

Antony Van Couvering
CEO, Minds + Machines