[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[registrars] NSI - Registrar contract comments


Hello, I was introduced to you in Berlin at the ICANN meetings,
but we did not have any in depth conversation.  My name is Richard
Lindsay, and I am the representative for registrar activities, 
of interQ Inc. one of the ICANN accredited  post testbed Registrars.

The purpose of this mail is to provide input to the current
NSI Registrar contract.  Rather than commenting on the exact
language, I would like to point out the broad areas that 
we feel need adjusting.  I am not a lawyer, but my overall
impression is that most honest lawyers would recommend 
that NO company sign the contract as is.  As far as I can
tell there is 100% liability on the Registrar, with 0%
control over the circumstances that could cause liability.
On the other hand, NSI has 0% of the liability, while it
retains 100% of the control over the circumstances that 
could cause liability.

Specifically, interQ finds the following to be very disturbing:

The $100,000 Performance Bond:

First of all, this is not at all a common practice in 
Japan.  While there are probably ways to accomplish
the same purpose, it will require negotiations with a
bank.  A performance bond with the recipient being a foreign
company further complicates matters.  Also, the only way 
to set up such a bond will require the applicant to have
sufficient cash reserves (basically the amount of the 
bond) reserved for this bond.  Given the current economic
situation in Japan, banks are extremely reluctant to 
finance any small or medium sized companies, and would
likely refuse to issue such a performance bond.

This requirement then would preclude ANY small Registrars 
from operating in Japan.  Most likely the only Japanese
companies that would qualify for such a bond would be 
extremely large corporations (Sony, NTT, Fujitsu, NEC, etc.)
I do not think this is the intention of the ICANN process.

The $9 Registry Fee:  

Considering that the current model is one of an extremely 
light Registry, and a heavy Registrar, this fee is 
exorbitantly high.  In effect, as a service, NSI is 
only conducting one data entry, is running the Registry 
DB, and operating the SRS.  Having run databases and 
systems of a much smaller scale, I would estimate the 
cost of each data entry is nowhere near this figure.  
I mentioned this to Chuck Gomez from NSI, and his
response was, in effect, that their system is so large, 
it cost a lot.  Of course it is true, large systems cost
a lot, but the larger they get, typically the more efficient
they get, and the less the cost of each transaction.

I would estimate the bulk of NSI's operational investments
have gone into the Registrar side; the customer support systems,
the billing systems, and all the supporting interfaces.
I have heard that the Department of Commerce has not officially
been able to conduct a full evaluation of NSI's operations and
a thus calculate a proper assessment of the true cost of each 
registration.  I would welcome DoC staff to examine interQ's
operations in Japan, where everything is much more expensive,
to use as a comparative model.  I would estimate the cost per
each registration would definitely not be above $2.00 USD, even
in Japan.

Transfer Policy for NSI Domains

While I can understand the concept of a fee for the transfer
transaction between Registrars, there is no rational reason 
to impose a new contract.  This particular policy is clearly
meant to retain domain registrations with NSI, and penalize
those who wish to obtain registration services from other 
Registrars.  This is clearly in contrast with the conditions
of Article No. 11.

Lack of a Standard Dispute Resolution Policy:

Although this is not directly related to the NSI contract, I 
feel that leaving the Dispute Resolution Policy completely 
up to the Registrar a shortsighted decision at best.  Besides
the fact that this sets the Registry/Registrar system up 
for easy sabotage (a Registrar could set a Dispute Resolution
Policy that invites controversy and non resolved disputes)
I do not find it realistic.  The Dispute Resolution Policy
should be a standard tool that the Registrar can turn to
in case of need, rather than an adventure in entering a land
of legal risk.  

I believe it is ICANN's responsibility to determine a fair
and reasonable Dispute Resolution Policy (with input from 
the Internet community) and establish this as a requirement
for operating as an ICANN accredited Registrar.  Recommending
all or portions of the WIPO report on one hand, and not
establishing a standard Dispute Resolution Policy on the
other seems to me a double standard.

Testbed Process and NSI SRS System:

Perhaps I am naive, but I fail to see the proprietary nature 
of running a gTLD Registry.  I would strongly encourage the 
Department of Commerce to put pressure on NSI to take the 
immediate steps:

1.  Tear up the NDAs restricting flow of data concerning 
the testbed Registrars and immediately open the process to 
public, and DoC scrutiny.
2.  Release all SRS source code and documentation to the IETF
to have the system go through the RFC process.
3.  Allow an independent technical review of the software
for security and performance evaluation.
4.  Immediately disclose accounting data such that an 
accurate assessment of transaction cost can be established.

Overall, I feel the Department of Commerce has a huge responsibility
to the Internet community as a whole, to ensure that the
decentralization and internationalization of the gTLD Registry -
Registrar system succeeds.  I do not feel that the current
NSI proposed Registrar contract, and the testbed process,
reflects the spirit of this whole process.  In fact, I believe
there is a strong divergence between the current process and 
the conditions stated in Amendment No. 11.

I appreciate your taking the time to read this rather long
email, and welcome any questions or comments.

Thank you,
_/_/_/interQ Incorporated
_/_/_/System Division
_/_/_/Director and General Manager
_/_/_/Richard A. S. Lindsay