From: Philippe Verdy (verdy_p@wanadoo.fr)
Date: Fri Jan 25 2008 - 07:02:22 CST
Doug Ewell wrote:
> Envoyé : vendredi 25 janvier 2008 09:14
> À : Unicode Mailing List
> Objet : Re: Questionable definition of Unicode
>
> Marion Gunn <mgunn at egt dot ie> replied to Jukka Korpela:
>
> >> I would strongly recommend against using the phrase "character set"
> >> at all,
> >
> > I'd warmly welcome your advice as to which term you currently
> > recommend instead of 'character set', Jukka.
>
> I'm neither Jukka nor a member of the UTC, but my advice would be to go
> with "coded character set" and be done with it.
If you just keep that, you're just missing almost all of what really makes
the Unicode standard, because this is just what the other standard, ISO
10646, defines, but not Unicode itself.
You need some other words to describe the fact that Unicode also maps a lot
of properties on top of this mappings, some of which are normative, some
other being informative and also being just a proposed correct practice for
its use (in 10646, there's also an informative property, shared by Unicode,
the representative glyph).
The normative properties mapped on characters also define strong
characteristic for the identity of characters, it helps differentiating
them, when the normative ISO 10646 mapping from character names to integer
is too much abstract or if it could create too many interoperability
problems when handling text.
Unicode is what makes the 10646 standard usable and interoperable in
practice for representing text. Unicode goes a little further by also
defining or documenting algorithms to handle, interpret, and modify text
properly while also keeping most of its meaning (something that ISO 10646
does not handle at all).
Without all these additions, the ISO 10646 "coded character set" would also
appear as extremely redundant.
This archive was generated by hypermail 2.1.5 : Fri Jan 25 2008 - 09:44:29 CST