At 15:57 09/06/2011, Nicolas Adam wrote:
>Dear JFC,
>Thx for reminding us that, not only alternatives may have merits on 
>their own, but that they also bear political impact.

Nicolas,

the difficulty (hence my more matter of fact subject) is to make 
people accept the possible impact of the technology evolution on the 
evolution of their daily use of technology use. RFC 5890-5895 and IAB 
RFC 6055 technically examplify (in the naming case, simpler to 
understand to GNSO members) that subisdiarity is the way the internet 
architecture supports large diversities. They look complex, very 
technical, etc. but in practical GNSO/ICANN terms they imply that the 
single root is to be virtual, i.e. co-manageable by billions of 
physical root files similar to the ICANNNTIA one.

This is not something new or that anyone can change. This is one of 
the three basic principles of the Internet architecture: adaptability 
(RFC 1958 - in 1998) of the large systems, simplicity (RFC 3439 - 
2020) of the very large systems and subsidiarity (these RFCs - in 
2010) of the very very large systems.

This means that core functions becomes weak if they keep being 
(de)centralized in front of large diversities that actually need 
pervasive coherent local responsibility. This means that the current 
ICANN is inadequate to the Internet TLD diversity they want to 
deploy. In selling their first new gTLD they will commit suicide. And 
havoc in the internet.

This results from a fundamental flaw in the ICANN's conception. This 
flaw is a commercial misreading of subsidiarity as to "foster 
competition" while technically subsidiarity is the mutual respect of 
everyone's own reponsibility. ICANN selling TLDs is like 
Einsteing  deciding to agree with Ptolemy. ICANN is incompatible with 
the Internet size they want to trigger.

This time there will be no alternate roots.

- for speed, security, reliability, empowerment reasons I do not use 
any root server for years. On Windows. Intelligent users will do like me.
- Google will start helping managing and supporting "root names" (the 
name Bob Tréhin gave to ccTLDs in 1977).

Please remember Vint Cerf, Google Internet VP, was the Chair of the 
WG/IDNAbis which worked out these RFCs I was eventually able to 
support after a ten years dispute to get them.
Please remember that ICANN was created because Jon Postel tried what 
these RFCs lead to. It was pure Internet legacy architectural 
visionnary management. Only the ICANN concept could stop it, and it 
did for a decade. Until the ICANN Australian Naming Team decided to 
sell TLDs M$ 250 a piece.

Cheers.

jfc
facilitator, [log in to unmask]