Re: [idn] opting out of SC/TC equivalence

["tsenglm@????.??.tw" <tsenglm@cc.ncu.edu.tw>]

----- Original Message -----
From: "Patrik Faltstrom" <paf@cisco.com>
To: "tsenglm@計網中心.中大.tw" <tsenglm@cc.ncu.edu.tw>; "Harald Tveit
Alvestrand" <harald@alvestrand.no>; <liana.ydisg@juno.com>; "hlqian"
<hlqian@cnnic.net.cn>
Cc: <huangk@alum.sinica.edu>; <idn@ops.ietf.org>
Sent: Saturday, September 01, 2001 4:40 AM
Subject: Re: [idn] opting out of SC/TC equivalence


> --On 2001-09-01 00.44 +0800
> "=?utf-8?B?dHNlbmdsbUDoqIjntrLkuK3lv4Mu5Lit5aSnLnR3?="
> <tsenglm@cc.ncu.edu.tw> wrote:
>
> >            As Patrik F?瞻ltstr?繞m  described , if it is a mechanism in
> > UNICODE consosium , then they have the version update procedures and
IDNA
> > also treat  the version update in UNICODE , why it can not be updated
step
> > by step  to  increse it ?
>
> Very simple answer.
>
> Let's assume that the table day one include the following characters:
>
> A
> B
> C
> D
> E
> F
>
> We agree that the simple equivalence rules for 1-1 mapping maps A->B.
>
> This means that the following characters are available for domain name
> registration:
>
> B
> C
> D
> E
> F
>
> People register domainames such as:
>
> B.example.com.
> C.example.com.
> D.example.com.
> E.example.com.
> F.example.com.
>
> Five different people register these five domains, which are different.
>
> If one type in A.example.com in the browser, nameprep will map that name
to
> B.example.com. So, A.example.com and B.example.com are treated as "the
same
> domain".
>
> Now, let's say that we increase the size of that table, and decide that D
> and E are equivalent, so D->E mapping should be added.
>
> This means that only the following characters are allowed:
>
> B
> C
> E
> F
>
> The domains D.example.com and E.example.com are equivalent.
>
> Now the questions, and I want you to answer them. Short answers are
> required:
>
> (1) Should the holder of D.example.com or the holder of E.example.com be
> forced to give back his domain?
>
> (2) Who is making that descision?
>
> (3) Who will tell the domain name holder that he is not allowed to have
his
> domain anymore?
>

          This is a very good example we all concerned . I answer your
problem by your
proposal (you  tell  Prof. Ho in another mail about the version update and
control). In  draft-ietf-idn-nameprep-05.txt :

A subsequent version of this document that references a newer version of
ISO/IEC 10646 with new code points will inherently have some code points
move from category U to either D, MN, or AO. For backwards
compatibility, no future version of this document will move code points
from any other category. That is, no current AO, MN, or D code points
will ever change to a different category.

          The approach assign the data set in some states,  From created
state to final fixed state , it is transited in one direction way on the
state diagram, it can not back and loop if you want the results are backward
compatible.   That is say if you want to keep backward compatible , you can
differentiate it step by step but you can not merge them further except you
can force the conflict one disappear.  Our language teacher always tell us
TC is always compatile to old age, that can explain why  there are so many
characters in chinese.
          By this principle,  why partial set of CJK characters that are
partial setted by local language tag can be used with different TLD
(cn,jp,tw..). Because the TLD implied the language tag and will let them
differentiable from other TLD .  The problems com from these contraints:
          1. Big table of UNICODE let domain name can be mixing registed but
viewed in part now.
          2. Many duplicate, easy-to-confuse and
un-normalied/non-canonicalied scripts in UNICODE.
          3. No one can change the coding at one night , you can only
transit step by step.
          4. ML.com try to use all code point withou considering the
troubles come from mixing.

L.M.Tseng