[vip] [ncsg-policy] Proposed NCUC Comments on the WHOIS Review Team Discussion Paper

[JFC Morfin]

At 12:17 22/07/2011, Timothe Litt wrote:
>Like driving, a network presence, including a domain name, is a 
>privilege and not an absolute right.


Who decided that? Milton has addressed that point.
However, Milton reintroduces the point when he states: " Instead we 
put limits on who can access this database (the police, LEAs) and the 
uses to which the data can be put." Who is "we". There is a single 
authority: the zone manager, there is only one law: the national law 
of the zone manager. There is only one rule to respect: the most 
stringent sovereign privacy rule, worldwide - otherwise there is not 
world of right.


Now, let me clarify.
There is no right or privilege in Internet use, there are facts. 
Rights and privileges may only concern Internet usages and people's 
Internet related behaviours and be enforced by governance regalian entities.


The Internet is a technical consensus
It works the way its programs are written. Programs are written to 
work. To obtain it, developers consider RFCs in the OSI layers 1 to 
7; and listen to the users (the "market") otherwise. This is why the 
Internet constitution is in the code, not in the ICANN community. 
ICANN is the leader of one of the communities populating the Internet 
community, which is actually a community of communities.


In the sole naming area, the Internet community has :
- one single rule which is the DNS that provide information enough 
(mail, nameserver, registry, zone manager) to be maintained.
- at least seven sub-communities that technology MUST support:

1. ICANN full-rate gTLDs, the vip at icann.org mailing list should 
discuss the technical requirements.
2. ICANN JAS-rated. There is no indication yet about their possible 
technical difference with the above.
3. open source gTLDs.
4. government created gTLD, e.g. China.
5. ISO 3166/MA decided ccTLDs.
6. industrial community (GSMA, Google, etc.) to support their own 
root. Who knows about .gsm?
7. non-Internet limited emergent IUse community and IUI (Intelligent 
Use Interface) related worked-on technologies. IDNA2008 exemplified 
how RFCs fully support it.


ICANN has documented this situation.
That was through its 2001 ICP-3, 
<http://www.icann.org/en/icp/icp-3.htm>http://www.icann.org/en/icp/icp-3.htm 
that states: "In an ever-evolving Internet, ultimately there may be 
better architectures for getting the job done where the need for a 
single, authoritative root will not be an issue. But that is not the 
case today. And the transition to such an architecture, should it 
emerge, would require community-based approaches. In the interim, 
responsible experimentation should be encouraged".


In 2011, in approving the gTLDs system,
ICANN has acknowledged that what was not the case in 2001 is in fact 
the case today. The single, authoritative root is not a limited file 
anymore that is disseminated by the root server system. ICANN tends 
to present it as an open file while IUI sees it as a virtual matrix 
with millions of dimensions. IUI responds to the WSIS demand for a 
people centered society and inherits from a community experience that 
was acquired along the ICP-3 rules ("dot-root" project): everyone 
runs and creates his/her own needed part of a root that is not 
limited to names of the sole Internet. This results from a third 
principle in the Internet architecture (RFC 1958: permanent change; 
RFC 3439: simplicity) the principle of subsidiarity that IDNA2008 exemplifies.


IDNA2008 was a positive surprise.
The IDNA2008 positive surprise was that subsidiarity is built-in the 
Internet architecture from the very begining. There is not a single 
bit to change. However, what has not been done yet is to document the 
transition to get rid of the unnecessary added-complexity that has 
accumulated over the decades and to welcome the permitted innovation. 
This is the challenge: to simultaneously support and test opposed 
transitions like ICANN and IUse (and what is in between) without 
confusing either of them. This is why both of them have engaged in an 
analysis and documentation efforts and should try to cooperate (e.g. 
http://idna2010)


This is why we need a consensual cooperation.
However, this cooperation is to be engaged in a technically and 
politically confused context due to the complementary charters of 
IETF and of the UNICODE Consortium, and to the discrepancies between 
the GAC and the WSIS objectives.

1. Clarification was obtained last year from IESG and IAB through my 
appeal over the IESG misrepresentation of the importance of IDNA2008. 
I could summarize it as: the IUI is an interface between the Internet 
(and alternate network technologies) and the external world. IETF is 
competent and interested in what belongs to or impacts the Internet 
but does not want to engage outside of the Internet area.

2. A clarification should be found with UNICODE through the 
"stringprep" replacement. IDNA2003 used "stringprep" to interface 
UNICODE to punycode entries. Stringprep is used by other IETF 
protocols but turns obsolete, since IDNA2008 does not use it anymore, 
freeing IDNA from Unicode versioning. The IETF (WG/PRECIS) tries to 
work out a solution. For good reasons that have endangered the 
IDNA2008 consensus, IUsers have a different vision of Unicode 
deliverables, and would like to consider starting at a deeper point 
of simplicity.


IDNA2008 has protected IDNA from Unicode versioning.
IUsers would like the IUI to protect them from Unicode and to void 
the need for stringprep in considering a no-phishing network protocol 
oriented scripting based on the visual aspects of the characters 
symbols in an unique common fount. This would remove the Unicode 
consortium from the network multilingualization loop and ease the 
naming adminance (long-term netkeeping, as opposed to medium-term 
governance and short-term operance).


Two systems are actually of no real technical and operational use in 
well organized and secure DNS operations:
- the root server system that answers 96% of erroneous requests and 
data everyone already has.
- the WHOIS is system that violates the privacy law of most of the 
countries having one - and is a source for spaming, spoofing, etc.

In addition, one system will become local and needs to get reviewed 
now ICANN has started selling $ 185.000 + expenses what everyone can 
deploy for free and get used by billions (reasonably within less than 
five years due to existing RFCs, word of the mouth, testing, new 
products and services being supported) or Google can deploy in minutes.
That system is the ICANN whole technico-legal system itself.


This means that the priority is to correctly insert that ICANN system 
into the foreseeable future development of the world digital ecosystem (WDE).
This is to protect stable operations and usage by its customers. This 
cannot be done by rules or agreements (you cannot negotiate with 
billions of individuals). It can only be done through:
- a stable, secure, simple, innovative technology these individuals 
will want to use for free as a "Plus" to the Internet they are used 
to utilize everyday. For many reasons IAB started to document 
IDNinApplication and xNAMES in the DNS as we consider them today are 
inadequate. However, a reponse MUST be found.
- and support services they will competitively adopt. I doubt the 
WHOIS is going to be a major part of such services: because it does 
not propose anything to the advantage of the registrant. Ths WHOIS 
only was a Jon Postel's tool to manage "his" network with people 
moving around every academic year. A dinosaur.

What actually IDNA2008 says is: here the way for DNS oriented 
concerns to interface the Internet DNS. The rest is to be entirely 
reviewed accordingly.

Best
jfc

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mm.icann.org/pipermail/vip/attachments/20110723/b63dd43e/attachment.html