Comments on CRAI Study

["Michael Palage" <michael@xxxxxxxxxx>]

I respectfully submit the following comments in an individual capacity and
not on behalf of any past, current or future client(s). In the interest of
openness and transparency it is important to note that I provide consulting
and management services to domain name registration authorities, and in the
past have advocated on behalf of multiple sponsored TLD clients for a
relaxation of the registrar/registry separation requirement. 

The CRAI report should be commended for its meticulous analysis of the
evolution of the current registry/registrar dichotomy, however, its
recommendation for relaxing the existing restrictions are vague/ambiguous
and without further clarification will impede the pending new gTLD process.
Failure to properly address these shortcomings now will result in the ICANN
Board having to make these decisions during the new gTLD process in an ad
hoc manner which benefits no one.

To illustrate just a small handful of the potential ambiguous scenarios
consider the following: 

Scenario #1 . Proxy Domains R US (PDRU) applies for a TLD .PROXY. All of the
domain names are legally registered to PDRU although the ?beneficial? owner
of the domain name is a third party registrant.   PDRU is a for profit
company.

Result #1: Does this fail because the CRAI study states that ?TLDs that are
being operated as a money-making venture should be excluded from this
category??


Scenario #2: Proxy Domains R US (PDRU) creates a non-profit subsidiary
PDRU-TLD.  PDRU-TLD gives away domains names in .PROXY to PDRU customers.

Result #2: Does this fail because the non-profit PRDU-TLD is the registry
and the for-profit PRDU is the registrant are not one and the same?


Scenario #3. The Privacy Advocacy Group (PAG) is a non-profit organization
established to promote individual privacy and free speech. They apply for 
the TLD .FreeSpeech. PAG registers all second level domain names in its name
for privacy reasons , although the ?beneficial? owner of the domain name is
a third party. 

Result #3: Does this pass because PAG is non-profit, and the registry (PAG)
is the same as the  registrant (PAG)?


Scenario #4. The Privacy Advocacy Group (PAG) is a non-profit organization
established to promote individual privacy and free speech. They apply for 
the TLD .FreeSpeech. PAG intends to register all second level domain names
in its name for privacy reasons , although the ?beneficial? owner of the
domain name is a third party. PAG decides to turn to DNS Infrastructure
Company (DNSIC) to provide back end technical and administrative services
for both the Registry and Registrar in order to meet ICANN technical
requirement criteria. Over 80% of the fees collected by PAG as part of its
not-for-profit service is paid to DNSIC for the services provided.

Result #4: Does this pass because PAG is a non-profit, and the registry
(PAG) is the same as the registrant (PAG), or does the fact that DNSIC is
the ?beneficial? recipient of over 80% of the fees have any impact on
whether this qualifies under CRAI exemption #1?


Scenario #4(a): Same as #4, except, DNSIC receives 51% of the fees collected
by PAG?


Scenario #4(b): Same as #4, except, DNSIC receives 49% of the fees collected
by PAG?


Scenario #4(c): Same as #4, except, DNSIC receives 33% of the fees collected
by PAG?

If permitted additional time, one could come up with numerous more
hypotheticals which could advocate both in favor and against the relaxation
of the existing registrar/registry separation requirements.


As I have stated in the past the registry/registrar dichotomy has served an
important role in promoting innovation and competition within the domain
name industry, and there should be a continued strong presumption in favor
of this distribution model in the near foreseeable future. However, this
presumption should be a rebuttable one in which a registry can demonstrate
how deviation/relaxation of the existing separations requirements will not
negatively impact consumer choice, innovation, and competition.  One
objective criteria that I originally advocated as far back as 2001 during
the negotiation of the original sponsored TLD agreements was permitting a
registry to provide domain name registration services direct to registrants
provided that registration volumes did not exceed a certain threshold level,
i.e. 50,000.  This approach was finally recognized by ICANN in connection
with the renewal of the .MUSEUM registry agreement, although at a much
smaller threshold level.

ICANN has two choices:  (Option A) ICANN can move forward with the new gTLD
process with a total lack of clarity involving this important component,
thus requiring the ICANN Board to make decisions as they arise during the
application process or (Option B) ICANN can invest additional time and
resources to provide the clarity, predictability and equality that the ICANN
community need and demand. 

Assuming ICANN opts for Option B, It is recommended that the ICANN staff in
consultation with the community, including potential applicants, create a
list of hypotheticals that should be provided to CRAI for clarification as
to whether the proposed registry business model qualifies for a relaxation
of the existing registry/registrar requirement. 

Sincerely,

Michael D. Palage