Please note that, in the current draft proposal for ISO
10646-2, plane 3 is reserved for CJK extensions. Eventually,
this should contain most of the characters under discussion,
but I wouldn't recommend making individual use of this plane
for Han characters outside of the context of an existing draft
proposal for such characters.
Since there is a need specifically not to unify the characters,
it may be necessary to rely heavily on private use characters.
Why not do all that is needed in planes 15 and 16, with any
additional character needs in the PUA of the BMP. That gives a
total of 137,072 characters, not counting those already in
Unicode 2.1 and the additional characters to be added in 3.0.
If, indeed, that is not enough space, you might want to pursue
a non-standard-conformant course of action, but I would suggest
that this begin with plane 13 (plane 14 is already being used)
and moving down as needed. If you do so, however, bear in mind
that, at any time, characters could be given standard
allocation in any of planes 1 through 15.
Peter
if indeed more than 128k private characters are to be used
here, and utf-16 is the encoding of choice in a system that is
otherwise ignoring common unicode encoding practices like
unification, then why don't you use other, unassigned planes
that are accessible with utf-16? planes 3, 4, 5, 6, 7, 8, 9,
10, 11, 12, 13 will likely go unused for some time. that yields
704k characters more...
markus
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:45 EDT