From: Hans Aberg (haberg@math.su.se)
Date: Thu Apr 14 2005 - 12:54:29 CST
I can note that if every Unicode abstract character has a unique,
unchanging, character name, then that can be used to deepen the
analysis of these characters, freed from the character numbering.
This can be used to develop new character sets, which take advantage
of the lessons derived from the original Unicode character set, but
also new computer technologies, such as smart rendering, the latter
which might make it unnecessary to have duplicate single characters
representing common combined characters. A problem with the original
Unicode set, is that the abstract characters represent different
semantic qualities, mixed together in one bag. I have been able to
identify the classes linguistically semantic characters, rendering
characters, and also input characters; there may be more. This
picture is not at all as controversial, as some of folks on this list
imagine it to be, because the original Unicode character set will
move on and take its place as a good compromise engineering solution,
admitting a major part of the human writings writable in computers.
It will also require a great deal of insight, and therefore time, to
develop new character sets along the lines I have indicated, plus
intelligent computer software to handle it. So the practical use of
such new character sets lie further down in the future. It would be
nice though, to see it happen within an open-minded extension of the
Unicode project.
-- Hans Aberg
This archive was generated by hypermail 2.1.5 : Thu Apr 14 2005 - 13:08:34 CST