> -----Original Message-----
> From: James Seng [mailto:jseng@pobox.org.sg]
> Sent: Tuesday, June 13, 2000 2:13 AM
> To: Maurizio Codogno
> Cc: idn@ops.ietf.org
> Subject: Re: [idn] case folding
>
>
> Maurizio Codogno wrote:
> > This won't work for ideograms, probably, but we already excluded
> > them in this discussion.
>
> Actually it does, in Unicode v3.0.
>
> It is possible to compose ideograms in Unicode v3 with few
> basic composition.
I guess you are referring to the 'ideographic description' characters.
They are (normally) visible characters intended to be used to
*describe* (usually unencoded, as yet, but possibly also already
encoded) ideographic characters. There is *no* requirement that a
renderer actually dynamically composes an appropriate glyph for the
ideograph *described*, and the ideographic description characters have
no implication at all for normalisation, as there is no (canonical or
compatibility) decomposition of ideographic characters into 'ideographic
description sequences'. I you want to make such decompositions in some
way, you're on your own.
An issue related to this is the 41 000 or so ideographic characters
that will go into plane 2. (I have no idea whether a 'Rong' ideograph
is there or not. I'm happy if I can recognise zero, one, two, three,
and ten when written as ideographs...)
Kind regards
/kent k
> The idea is that we can have ideograms which is not currently
> defined in the
> CJK U+4E00 range. (e.g. Zhu Rong ji - the 'rong' is undefined).
>
> While basically this is a good idea, I like to see the face
> of the guy working
> on the normalization of hans character. :)
>
> -James Seng
>