At 05:36 03/03/2006, Mark Andrews wrote:
They are still a problem whether you think they should exist
or not. The problem is that they are added unilaterally
and people using them expect everyone else to be able to
resolve them as well. The method of adding them was wrong
as it does not scale. If every language added the equivalent
you would have hundreds of sets of nameservers that you
would have to track down and add to your own configuration.
This does not express a problem in itself. This expresses a 22 years
old need. The IETF problem should be to address that need. But it
seems this is not a problem.
The whole point of the DNS was that you didn't need to do
that because it provide a *single* namespace from a *single*
set of servers and you didn't have to graft on hundreds of
TLDs.
This is not a DNS feature. This is a result of the unique space of
the prototype network. This is precisely because the Internet
technology cannot (yet) support multiple spaces that the Chinese
solution is not off the shelves. But documenting this solution
(permitting externets consistency, continuity and interoperability)
seems not to be the IETF's cup of tea. I approached the problem in
different ways (open roots, network architecture parameters, ICANN's
ICP-3 call for a test-bed in that domain, liaison with ccTLDs,
multilingualism, multilateralism, IPv6, ethic, etc). The IETF
consensus is consistent: no real interest, predetermined doctrine,
some active confusion serving commercial interests (as documented by
IAB in RFC 3869) and cultural hysteresis (as documented by RFC 3744).
I conclude this is now to be documented by a TF specialising in
International Network architecture and transnetwork/transtechnology
solutions. This calls for a generalised vision wich will probably
better develop within the IGF. In the increasing number of people I
know sharing that evaluation I am currently alone (and opposed on
both sides) actively supporting that interoperability with IETF
propositions should be maintained as much as possible. Or the lack of
interoperability to be limited (this is what I succeded to do with
RFC 3066 Bis). A solution is to keep informing the IETF through IETF
Drafts and IESG appeals when interoperability/interintelligibility
seem compromised. Another is to organise a structure assuming liaison
as part of the Internet standard process. I documented that project
as the IGFTF in my positive (on other grounds) appeal to IAB. I
however delayed it to permit Brian to possibly update RFC 2026 such a
project should strictly respect.
jfc
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf