Re: Thread on - Re: [idn] Prohibit CDN code points

[Patrik Fältström <paf@cisco.com>]

--On 2002-01-26 20.59 +0800 "=?big5?B?dHNlbmdsbUCtcLr0pKSk3y6kpKRqLnR3?="
<tsenglm@cc.ncu.edu.tw> wrote:

>           You just point out the problem ,  how to differentiate them ?
> IDNA  module is a common module shared by most applications or a full set
> belong to each AP  independetly is not the same results in these
> descriptions .  Does some one can clarify it more clearly ?

IDNA specifies what goes on the wire between two applications. Not what
libraries, functions etc are to be developed to reach that goal.

The functions specified in the IDNA draft is a functional specification on
what must happen with the strings. Not a specification of functions which
should exist in libc or such.

My personal guess is that we will see ToUnicode and ToAscii functions in
libraries which already take care of charset conversion. Like the Text
Encoding Converter from Apple. (see
http://developer.apple.com/techpubs/macos8/TextIntlSvcs/TextEncodingConvers
ionManager/TEC1.5/TEC.38.html)

Similar modules will certainly be developed as perl libraries and what not.

Many of them probably takes a native charset as input (such as BIG-5 or
ISO-8859-1) and not some encoding of Unicode.

IETF extremely rarely develops API's.

  paf