[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [idn] Mapping and Prohibit of code points



Dear  Soobok:
          I  agree your point .  We may forget current Punycode only
supports single case per character. Even there are a higher service layer
above DNS, the multiple matching result will not apply  the good properties
of multiple-case ACE encoder to connect to 7 bits LDH  DNS server .
          Although there are a lot of equivalent character problems in
UNICODE  can not be solved  quickly,  but the fixed monolithtic IDN approach
should include the flexibility of multi-cased Punycode to support the
furture connection for higher layer service.

L.N.Tseng
----- Original Message -----
From: Postel Services, Inc
....
>In order to do multi-case Punycode correctly and suffiencienly  for TC/SC
and >latin/cherokee/gree/cyrillic and hangul jamo matters and turkish i and
others,  we >should devise *new layer* that would replace current stringprep
one. stringprep is far >from what users want.
...
>In summary,  multi-case Punycode require  completely new reworks of IDNA
draft sets. >it will be like making new ones. I believe it will take another
2~3 years..
...
>Personally, i believe that  DNS should  remain as 7bit ASCII one which is
>internationally recognizable and distiguinshable and verifiable. Instead,
the users' need >for "internationalized *access* to domain names" should  be
solved  *above DNS*   by >the best ways that can serve local user
communities. It will bring  users   more benefits >and conveniences than
fixed-forever monolithic  IDN approach.