[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [idn] Chinese Domain Name Consortium (CDNC) Declaration
Dear Paul Hoffman:
> At 2:23 AM +0800 2/10/02, tsenglm@cc.ncu.edu.tw wrote:
> >Can you specify more clearly ? we all like to know what is the reference
?
>
> Sure. The "reference" is personal conversations that folks from CDNC
> had with other people active in the IDN WG during the Salt Lake City
> meeting. Many people met with CDNC members in Salt Lake City, and
> many of us heard similar statements in our discussions.
>
> > 1. make more money from registration
>
> People from both CNNIC and TWNIC have said "we cannot do a 2^n
> registration solution because we need to charge end users for each
> name in the zone".
Are you sure you do not make any mistake in these statements ?
In SLT , I have heard from Director . Mao
talk to John Klensin , making more money is not the goal of CNNIC , so he
can accept the delay of CJK deployment to protect all CJK users. CEO of
TWNIC Prof. W.S. Chen is not in SLT , I can make sure it is also the same
answer as CNNIC .
> Of course, you do not need to charge per name: you
> can charge per group of names that come from a single base name. But
> the fact that this argument still comes up leads one to think that
> maybe you want to charge per name even though doing that will hurt
> the Chinese people.
>
> > 2. it is impossible to solve TC/SC
>
> I never said that; I have said the opposite many times. What I said
> was that it is impossible to get the solutions proposed in the tsconv
> Internet Drafts approved by the IETF because of their obvious
> technical and political flaws, all of which have been openly
> discussed in the WG. During meetings in Salt Lake City, CDNC members
> agreed with this and said that they would pursue a good
> Traditional-Simplified solution outside the confines of the IETF.
>
Sorry , I do not see discussions of TC/SC related to
technique of Validation , HSE , Multi-Case ACE in here . I am also in SLT
, why I have not this solution ? We hope this WG temporarily delay the
CJK deployment by code point inhibition to let CJK area can get a TC/SC
solution if this WG is so hurry to pass all, because it is incomplete and
irreversible for user to register CJK name.
> > 3. registration policy can solve all ...
>
> No one has ever said that "registration policy can solve all ...".
> What has been said, and agreed to by TWNIC people, is that in the
> absence of Traditional-Simplifed mapping in IDN, registration policy
> can serve end users. It will not be consistent, so it will not serve
> them as well as it would if we could have put it into IDN. Of course,
> doing T-S registration can serve users better than anything that ever
> appeared in the tsconv drafts because it will not be limited to 1:1
> mapping, and it will allow mapping that comes from humans instead of
> limited tables.
>
I think people from TWNIC recognizes that to solve TC/SC
problems , it need multiple stage/level to cooperate . 1-1 mapping table
are proposed in the draft that can be applied in the mapping after nameprep
and before Puny code encoding. 1-n mapping need more time to discuss and
select one of them as 1-1 mapping or no any mapping , if no-mapping is
assigned by language expert , the registration approach can help to register
all of them or select few of them by register.
All the argument in here is " 1-1 mapping should be doing in
which level ?" 1. inside IDNA 2. Outside-IDNA: a client local module or
level-2 server like IRNSS and what is the connection interface between
this module with IDNA, 3. Registration policy .
L.M.Tseng