- From: Glenn Adams <glenn@spyglass.com>
- Date: Thu, 1 Feb 96 09:12:06 -0500
- To: Masataka Ohta <mohta@necom830.cc.titech.ac.jp>
- Cc: keld@dkuug.dk, masinter@parc.xerox.com, yergeau@alis.ca, Dan.Oscarsson@malmo.trab.se, html-wg@fssun09.dev.oclc.org, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, maits@dkuug.dk, uri@bunyip.com
From: Masataka Ohta <mohta@necom830.cc.titech.ac.jp>
Date: Thu, 1 Feb 96 10:30:01 JST
> ... for example big-5 is not covered by ISO 10646 ...
>
> Why do you say this?
In Big-5, two characters are duplicatedly encoded to have two code
points. So, mapping from CNS 11643 or ISO 10646 can not be one to
one.
Keld said "covered" not "maps one-to-one". In any case, I agree that
there exists a need to use other character sets with the %XX encoding.
On the other hand, it would be reasonable to indicate a default inter-
pretation, and, for this purpose, I would suggest the UTF-8 encoding of
ISO/IEC 10646 (= Unicode 1.1).
Regards,
Glenn
Received on Thursday, 1 February 1996 06:26:53 UTC