RE: [idn] Comparisons of the proposals

["McDonald, Ira" <imcdonald@sharplabs.com>]

Hi,

Amplifying what Harald said below, *retaining* the labelling
of non-UTF-8 charset with domain names escaping into numerous
application and infrastructure protocol interfaces is another
compelling reason not to allow local (non-UTF-8) charsets in
IDNs.

Cheers,
- Ira McDonald, consulting architect at Sharp Labs America
  High North Inc


-----Original Message-----
From: Harald Alvestrand [mailto:Harald@Alvestrand.no]
Sent: Sunday, March 26, 2000 4:55 PM
To: James Seng; Dan
Cc: idn@ops.ietf.org
Subject: Re: [idn] Comparisons of the proposals


At 18:16 26.03.00 +0800, James Seng wrote:
>The question is why not? We could easily say allow any charset registered
with
>IANA and tag it according to the number assigned by IANA. This allows both
>localized encodings and also UTF-8. We are also not forcing people to use
>Unicode (for those who really really hate Unicode).

The big problem with using multiple character sets is always what set of
character sets one requires a client to be able to handle intelligently.

I would prefer that I can tell implementors "if you do intelligent handling 
of ONE character set, you don't need to worry about others".
If I have to tell them "you have to do intelligent handling of 8859-2 in 
Poland, 8859-7 in Greece, EUC-KR in Korea, Shift-JIS in Japan, UTF-8 in 
Trondheim, and all of the above for the Internet", I don't feel happy.

Labelled noninteroperability is an improvement over just-send-8, but not a 
big one.

                      Harald

--
Harald Tveit Alvestrand, EDB Maxware, Norway
Harald.Alvestrand@edb.maxware.no