Note my comments below.
Chuck
-----Original Message-----
From: Ed Gerck [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, January 12, 2000 11:47 AM
To: [EMAIL PROTECTED]; [EMAIL PROTECTED];
[EMAIL PROTECTED]; [EMAIL PROTECTED]
Subject: [IFWP] Re: missing the dot
"Gomes, Chuck" wrote:
> The original Registrant is not out of luck. He/she has an agreement with
> the registrar containing specific terms and conditions that both the
> registrant and the registrar are required to follow. If either violates
> those terms and conditions, the other has legal recourse. How is that
> different than the way most businesses operate?
Because that agreement is heavy-handed in favor of the Registrant,
mandatory, uniform, precludes legal recourse, gives the ultimate authority
over the domain name to the Registrar -- and the registrant cannot choose
any other business since they are all the same. There is no competition
here, Chuck, when it comes to contract terms. The choice is to have
no choice.
This seems to be different than what you say in the last paragraph: "the
original registrant of a domain name has ownership of that name."
Further, most (legitimate) business do not conceal tracing the origin
of their product to the customer -- and this is what the Registrar does
under the Shared Registry system, since the customer cannot ever trace
the origin of the product s/he bought -- the Registry. The Registry will
always deny selling that product to the customer or having any
responsibility
whatsoever to the product -- even though the customer's identification
(the domain name) is in its files and points to the customer's host
addresses.
The Registry certainly would denying selling a product to a registrar
because it does not do that.
Further still, this is a business where the reseller (the Registrar) keeps
the
property of what it sells -- as I quoted before, Section II.H "Rights in
Data"
states that a Registrar has "the rights of an owner to the data elements
listed
in Sections II.E.1.d and e and II.F.1.d through i concerning that
registration".
Note that the domain name is not included in that list.
In essence, the difference is that the reseller (Registrar) follows a sales
system
(Shared Registry) that denies tracing the producer (Registry), denies
responsibility
to the buyer from the producer, denies access to producer logs in order to
prove
buyer's rights, keeps the property of what is sold with the reseller, gives
the
buyer no authority over what was bought, and denies legal recourse as the
buyer may see fit.
I have a hard time seeing the Registry as a producer. The manufacturing
analogy seems to be weak here.
Concerning luck, and under this kind of system, the registrant really needs
it ;-)
> Help me understand how a Registrar could change the registrant field to
> whatever they wanted without violating agreements with ICANN and/or the
> Registrant.
You are mixing apples with speedboats. Ask not what the Registrar can do
for ICANN, the matter here is the registrant. Let me rephrase your question
as:
"Help me understand how a Registrar could change the registrant field to
whatever they wanted without violating agreements with the Registrant."
Not sure I see much of a difference but I can accept this question.
Easy. They can say without proof that the Registry did not accept the
registration -- even though the registration is already paid, or already
renewed, by the customer. They can transfer the registration to another
Registrar (slamming) because they *own* the rights to that data.
So the registrant needs to obtain proof from the registrar.
And, they can surely change anything in the Registry with no recourse
from the registrant to the Registry since they are the ones responsible
for enforcing their own compliance to the Registry-Registrar agreement
in regard to domain names -- I know it sounds funny, but I already
quoted this from the NSI-Registrar agreement:
"Enforcement shall be the responsibility of the Registrar sponsoring the
domain name registration."
so, hands off for the Registry. The Registry is of no help if you use a
Registrar that the Registry has "accredited" (in the real world, to accredit
means "to believe") -- the belief of the Registry on the Registrar does
not transfer to the registrant. In other words, to the registrant the
Registrar
is *not* accredited by the Registry -- it does sound funny, no?
I agree with you that the Registry would be of no help in this situation
unless there is evidence of a violation of the Registry-Registrar agreement.
You are correct that the registrar is not accredited by the Registry but
rather by ICANN. We had originally suggested a model where ICANN would flow
through its requirements through the Registry to the registrar, but not was
rejected.
Now, what happens if you include ICANN in the picture? Nothing -- all
these actions do not violate the agreements with ICANN. And, they
never violated the agreement with the registrant. The Registrar is
clean. The registrant is out of luck.
What are we talking about here? probably a few days at most before the
registrant would have some confirmation that the registration was accepted.
Once the registrant has confirmation that the registration has been
accepted, there is now legal recourse.
> From a practical point of view, it's very hard for me
> to imagine a Registrar staying in business very long if it ignores its
> agreements with its customers.
You are missing the dot ;-) The Registrar never ignored *any* agreement
in all that I exemplified above. Neither with the registrant, neither with
the Registry, nor with ICANN. The registrant is the one being ignored,
truly, but s/he has nowhere else to go -- all options are uniformly bad.
Large corporations have a way out as I explained -- become a Registrar.
Smaller corporations and individuals must beware.
> Registrants certainly have a responsibility
> to make sure that they are doing business with reputable businesses, but
> that is true in all businesses.
The problem here is that the business itself is not reputable -- since
accountability exists only very tenuously, and only to the middle-man
(Registrar). Think of it for a moment -- reputation is closed linked to
accountability and thus by refraining from taking accountability, the
Registry ends up with no reputation to the registrant. But, since the
registrant ultimately depends on actions from the Registry, the
entire chain of trust is broken right at the source.
> What would be gained if the Registry is put in the middle of a dispute
> between a Registrar and a Registrant? Then there would be three parties
in
> the dispute adding a new layer of complication.
Wow! you just did away with the legal system, with arbitration, with
Public-Key CAs, with notaries, with witnesses and with God knows
what -- hey, even with God ;-)
That's quite a stretch!
Now, let me go back to civilization ;-) In an Internet protocol, as well
as in the commercial system, we always like to have at least three
parties, even though it may seem to add a new layer of complication.
The reason is to be found in the study of trust -- self-assertions cannot
induce trust. We need a trusted introducer to begin the transaction
-- which role would be played by the accrediting Registry and by
ICANN, and we need a trusted witness to confirm the transaction --
which role, again, would be played by the Registry and by ICANN.
Thus, the Registry has not one but *two* essential roles to play in
regard to the registrant -- the Registry acts as a trusted introducer to
the Registrar and as a trusted witness. Of course, the extent of that
trust can vary from Registry to Registry and from registrar to registrar
(i.e., it is intersubjective) but there is a log, there is a trace, there is
evidence that can be used.
I understand your point even if I don't agree.
And, in life as well as in chess, it is the threats that are not fulfilled
which control the game -- if I, as Registrar, know that my customer
can ask for the logs of the Registry and that there will be evidence
of all transactions in that log, with possible legal effects, then I
*will* think twice before even trying to cover up or fouling-up.
To contrast, if I *know* as I know today that there is no witness
to my actions then I may as well do and say whatever I please,
even if what I do is not what I say.
This is the type of argument that is often used to support government
regulation. I certainly believe there are cases where government regulation
may be needed, but to allow market mechanisms to work freely, I personally
believe it should be kept to the utmost minimum.
So ... you ask what would be gained -- everything. Both roles are
needed. To deny both roles as the Shared Registry system does at
its conceptual level (see above quotes) is to deny the very
functionality the registrant needs. There is no trusted introducer
(hands off for the Registry) and there is no trusted witness (ditto).
This may be very convenient and simple to the Registry -- and
no one questions that. However, who is buying and who is
selling? Who has the power and who is the relying-party?
The Registry cannot deny knowing what is in its own files,
the Registrar cannot be the enforcer of its own controls,
the registrant should not be demoted from domain name
holder to a licensee of its own name.
But, in the US, I hear that the 5th, 1st and 10th us district courts
have already ruled that the original registrant of a domain name,
has ownership of that name, not ICANN nor the Registry. I also
hear that this is old news. Why is the point still missed?
Cheers,
Ed Gerck
