Single Letter Domain Name Allocation

["Michael D. Palage" <Michael@xxxxxxxxxx>]

I would like to submit the following personal comments regarding the
allocation of single letter domains at the second level. To begin, I
would like to  analyze the specific wording contained in the relevant
GNSO policy recommendation:

We recommend that single letters and digits be released at the second
level in future gTLDs, and that those currently reserved in existing
gTLDs should be released. This release should be contingent upon the use
of appropriate allocation frameworks. MORE WORK MAY BE NEEDED. (Emphasis
added).

Reserved Names Work Group Dynamics:

As a participant within the Reserved Names Working Group, it is helpful
to shed a little light upon the genesis of this exact wording from my
perspective. There were a diverse range of interests that constructively
participated in the working group, and I consider it one of the more
positive ICANN experiences in which I have participated over the years.
During the working group's consultation, it was explained how ICANN
accredited registries had already used the existing PDP approved funnel
process to allocate two letter "reserve names". However, there were
other participants that appeared to favor a more regulatory (i.e.
one-size fits all) approach that would be mandated on all gTLD registry
providers
through a PDP. Through the leadership of the chair and the professional
and collegial attitude of the working group members, the final
recommendation as stated above was approved unanimously.  

Registry Allocation Innovation:

It has become clear to even a casual observer of the domain name
industry, that ICANN's limited expanse of the domain name space has not
yet created the desired level of competition as original expected.
Therefore, the primary challenge confronting new registry operators is
how to spur adoption and use of their TLD. Simply relying upon Sunrise
and traditional first come first serve (FCFS) models is not sufficient.
History has repeated shown that corporate entities are most likely to
defensively register their domain names and then redirect them to their
existing .com site. The other likelihood is that professional
registrants will figure how to game the particular allocation system of
a registry and the secure the domain and either resell it or park it to
generate PPC advertising. Unfortunately neither of these circumstances
actually help the registry/sponsor build its brand. 

Consistent with the competitive and innovative principles upon ICANN is
built, new registry operators have learned from its predecessors and are
adapting to find new ways to make sure they identify the best people to
help build the brand of the registry. Two recent examples worth noting
are the Premium Name allocation process of .MOBI, and the Pioneering
Process of .ASIA. In each of these examples the registries have used a
mix of RFP and auction processes to allocate domain names in an
equitable process, other than a traditional FCFS process. These
processes, particularly in connection with RFPs, empower the registry to
indentify registrants that are willing to make a commitment to the long
term success of a TLD. 

ICANN's HISTORICAL DEFERENCE TO REGISTRIES VIA THE FUNNEL REQUEST
PROCESS REGARDING RESERVED NAME ALLOCATION: 

Single letter domain names are reserved under the same legal provision
as two letter domain names in the current registry agreements. This is
relevant because ICANN to date has allowed several registries to
un-reserve and allocate two-letter reserved domain names through
numerous funnel requests.
It is important to note that at no time did the community seek to
mandate a one-size fits all approach, nor extract addition revenue from
these registries. In fact, there appears to be a diversity of the
different mechanisms proposed by the registries properly tailored to
their respective business needs. 


ALLOCATION OF A LIMITED RESOURCE:


In the announcement accompanying the commencement of this public forum,
ICANN stated that "[s]ince revenue will result from this allocation,
comments regarding the potential uses for this revenue are also
requested."
As an organization that is primarily tasked with the technical
coordination of the Internet's Name and Numbering system, I find it
somewhat troubling that ICANN feels that it is part of its mandate to
analyze "potential uses"
of additional revenue that a registry operator might recognize through
the normal operation of its business. More troubling is that some
commentators appear to have embraced this position under the auspices
that this is a limited resource.  

However, when one looks to the very real reality of the exhaustion of
The IPV-4 space, we see no action by ICANN or the community to decide how
money can be extracted through the RIR's business operations to fund
other ICANN initiatives above their existing contributions.
Unfortunately, the gTLD space continues to be the primary source (over
90%) of ICANN's continuingly expanding budget. If ICANN believes that it
needs more money to fund its continued growth, than it should look to
other non-gTLD sources as oppose to the gTLD sources that continue to
bear the financial burden of ICANN's operations.


.COM TAIL WAGGING GNSO POLICY DOG

Some of the other commentators talk about the potential financial
windfall that registry operators might recognize if they were permitted
to allocate these reserved names with no oversight or regulation.
However, the evidence associated with the allocation of two letter
reserved names by other registries have shown that not to be the case to
date. Although most people do not want to talk about it publicly, they
are just opposed to the idea of VeriSign pocketing the substantial
revenue (likely tens of millions of
dollars) associated with the potential auction of single letter second
level .COM domain names. Unfortunately, this is yet again another
example of how the .COM tail, wags the GNSO policy development dog.  The
reality of the situation is that the potential auction of highest single
letter .COM name would likely exceeding the total value of all other
non-legacy gTLDs single letter auctions combined. 

That is why I feel so strongly that registries should be able to use the
existing funnel process to determine what process works best for their
registry and/or community. A uniform one size fits all approach simply
does not work, and is not in the best interest of competition and
innovation. Through the funnel process, the ICANN community already has
built in
safeguards to ensure that the new proposed registry service is an
"appropriate allocation framework."