From: Pavel Adamek (pavel.adamek@ima.cz)
Date: Fri Mar 19 2004 - 11:47:19 EST
> > COMBINING C BEFORE
> > <H><COMBINING C BEFORE> = CH
>
> Shhh. It's not April 1 yet.
Of course I do not want to add this character to Unicode,
I was only thinking about possibilities.
The document
"An operational model for characters and glyphs"
says:
-----
Even within the content domain,
the nature of the character coding employed
for textual data affects the type or types of
processing to be performed on the data; no
single coding can optimize more than a few
such potential processes.
------
From the viewpoint of sorting,
the coding <H><COMBINING C BEFORE>
would be much better than
<C><COMBINING H AFTER>.
Also from the viewpoint of sorting and searching,
it would be better to have not separate characters
for LATIN CAPITAL LETTERs and LATIN SMALL LETTERs,
but case insensitive LATIN LETTERs
plus several zero-width characters like
BEGIN OF SENTENCE,
BEGIN OF PROPPER NAME,
BEGIN OF GERMAN SUBSTANTIVE,
BEGIN OF ENGLISH VERSE,
which would provide context for choosing of a capital or a small glyph.
P.A.
P.A.
This archive was generated by hypermail 2.1.5 : Fri Mar 19 2004 - 12:31:30 EST