[discuss] reality check

[JFC Morfin]

David,

Thank you for this. This is the first time that I have read a mature 
evaluation of the problem at hand, from a network real operations and 
management point of view and one that I can not only agree with, but 
that I can also work on.

>At 04:29 21/03/2014, DAVID JOHNSON wrote:
>The entire idea of Icann, all of its powers, comes from its 
>contracts with registries (and, up to now, the decision of root 
>servers to view its decisions re the contents of the authoritative 
>root as authoritative).
>
>When it was formed, there was a deal. Not only would NSI agree to 
>the written contract with the newly formed ICANN. It would agree to 
>be bound by future policies.

That deal was twofold. One with ICANN and one with the USG.
http://www.sec.gov/Archives/edgar/data/1030341/0000950133-98-003490.txt
http://www.ntia.doc.gov/legacy/ntiahome/domainname/proposals/DOCNSI100698.htm

This was because "the President directed the Secretary of Commerce to 
privatize the domain name system (DNS) in a manner that increases 
competition and facilitates international participation in its 
management." http://www.gpo.gov/fdsys/pkg/FR-1998-06-10/html/98-15392.htm.

He said privatize, not "civilize" (irt. Civil Society)!

The whole problem is that this very document is a lie by "class" 
omission. It states "The U.S. Government should end its role in the 
Internet number and name address system in a manner that ensures the 
stability of the Internet. The introduction of a new management 
system should not disrupt current operations or create competing root 
systems."

It has never stated, but it has been read and enforced by an 
unwritten status quo, that this new management system should disrupt 
the current unique authoritative root per multiple class DNS 
architecture, and as a result create a DNS market monopoly.

>But not any policy decreed by the ICANN board.
>Only those supported by a consensus among affected parties -- and 
>only those dealing with the operation of the domain name system. 
>(not content or use of the net)
>That was the deal. A remarkable commitment for a public company to make.

It was protected by the US consensual acceptance of the NTIA's lie by 
ommission. Question today: if the NTIA retires, what about NSI obligations?
Is NSI forbidden to replicate and extend its business to an "NSI" class?

>(As ICANN's own lawyers might say, how could a board of directors 
>turn it's duties over to some outside group?)
>Not a commitment any of the ccTLDs (who already had their 
>delegations) would make.

Mike Roberts certainly remembers Peter DeBlanc's obligation of 
resorting to "our nuclear arsenal" if he went too far. Keith Davidson 
reminds the project of several ccTLDs to start a ccTLD root. This is 
what China has already organized. This is what several armies have 
organized for their own operations.

This is exactly the situation today. Peace prevailed as Govs 
considered the NTIA's ICANN as a historically leading ccTLD. Now, as 
per Clinton's order, the USG retires and wants the other Govs to do 
the same. This raises two problems:

1) Govs are not subjects of the USG. This "detail" is the real reason 
for Sao Paulo, which should start the delegation of the Internet from 
Govs to non-Govs. Dubai showed that it is not the idea of a broad 
majority of States. Errare humanum est, diabolicum perseverare.

2) The new IG would be conducted by a mutually agreed (actually 
observed) capability among MS. This means that Sao Paulo is about 
normalizing TLDs as a private sector under ICANN operational, 
contractual, and financial rules (civil society cannot and does not 
want to meet them). These rules are built on the NTIA lie by class 
omission and, therefore, suppose that the whole Internet is the DNS 
class "IN", which it definitely is not.

If the NTIA retires, the politically enforced de facto monopoly of 
the NTIA's VGN dilutes in the VGN multitude, and we enter in the 
VGNICs competition era.
I suppose that Public DNS (8.8.8.8 and 8.8.4.4) will become a root to 
remember. The "unique authoritative root" is not defined by law or 
contract, but by use.

>ICANN has systematically eroded this deal by imposing contracts not 
>based on consensus ,and not dealing only with the sound operation of 
>the domain name system, on new registries.

Definitely. BTW, I think we do not really hear the ALAC on this list. 
Should I perhaps revive atlarge.org?

>It has done so by using its power over what goes into the root -- 
>based on deference from the USG. But that is not a legitimate use of 
>that power. In effect, an abuse of the the IANA function. The forces 
>leading to this are easily understood. The need to resist them just as obvious.

Amen. For govs, businesses and people. Not out of any political 
opposition, but because the NTIA retirement plan has no credible 
substitute to the trust in its lie by class omission. The US 
strategies are based on too many lies.

>The question is whether we will have a system in which all TLD 
>registries agree, by contract, to follow ICANN policy rules.
>If ICANN exists to make policies necessary to protect the net, 
>"issues the uniform resolution of which are necessary to assure the 
>stable and secure operation of the internet", how could they refuse?

That would only secure the ICANN organized functioning of class "IN". 
But it is likely that people and trade will move toward more 
attractive classes, on ad equation to their own needs. Let us take a 
simple example: the DNSA proposes a TLD enhanced cooperation over a 
"KID" class at one buck a year more with an ethics committee.

>And, if so, what will constrain the resulting power?

Competition among class administrators, i.e. VGNICs.

>I only know one answer:

You will note that your answer conforms to what we clearly identified 
with the IDNA2008 RFCs as the IETF/Internet architecture to deal with 
diversity.

>(1) rules must be supported by consensus among affected  parties (to 
>preserve subsidiarity, to give legitimacy, to foster compliance),

Thank you. The difficult thing is to identify the meaning of "Multi" 
and "Stakeholders". "Affected parties" is perfect and clear in every 
legal language.

>(2) rules must only be on topics where global agreement is necessary 
>to protect the stable, secure operation of the net.

Yes! You will note that this is exactly what Fadi does not want to 
discuss in Sao Paulo: topics. Keeping commenting principles keeps us 
away from reality: the smoke screen of the smoke screen. I have 
always qualified ICANN as the NTIA's smokescreen. Now we fully see 
that the NTIA was not the public interest executive but the private 
industry lobby's own smokescreen. By the President's order.

>On other topics, there are to many divergent views -- global rules 
>won't work and are not desirable. Without consensus, minority views 
>will be unfairly suppressed.

Actually, in a network, minority views cannot be fully suppressed, so 
they radicalize. The idea to call them cyber-terrorism (this is for 
the NSA to detect the keyword and read this mail) cannot work for long.

>We can agree that there should be a global forum for forging 
>consensus on policies that govern the domain name system, to make it 
>work for everyone.
>Let's focus on that.

Right. This is the DNSA. And it will be open to everyone. Including 
the civil society, the private sector, and the governments on an 
equal footing.

jfc