You suggested that the industry may not be ready for 32-bit
codes. I had not thought about that. Glyph IDs in TrueType
fonts are 16-bit, I believe, so that is an issue for
TrueType. Also, you're very right that numbers are much
less useful for working with glyphs than mneumonic strings.
I don't think so. There shouldn't be a relation between glyph IDs and
input code points at all in TTFs! Cmaps have been invented to
establish the needed relationship.
You also suggested that Unicode is adequate as a glyph
encoding for many purposes. Apparently, Adobe is thinking
along those lines. It is a serious concern for us if they
decide to carve up the private use area and try to
quasi-standardize it. We may need to make heavy use of PUA
for scripts that we work with that are not close to being
standardized in Unicode. What I have seen of what Adobe
wants to put into their are mostly glyph variants, which
don't need Unicode values if one has a smart rendering
system.
I don't see the problem. You can always populate the PUA with your
own scripts; you don't need to care of any of these assignments! If
you properly separate input encoding from glyph addressing, no
problems will occur. E.g., you can assign to code U+F012 your `foo'
glyph, and at the same time you access glyph `uniF012' mapped to
another code point.
The overlapping of the Private Use Area and the Corporate Use Subarea
is only a concern if you want to do a 1-to-1 Unicode->Glyph mapping
(e.g. for unknown glyphs or glyphs which can't be handled by your
rendering system) -- of course, it's a shame that even recent software
written by Apple, Adobe, or Microsoft still rely on the 1-to-1
Unicode->Glyph model...
Adobe has said that they're working with Microsoft on
OpenType (see the first quotation above). If the two of them
got OpenType right, Adobe wouldn't need to assign PUA values
to glyph variants. Unfortunately, Adobe and MS appear not to
have had much success in figuring out complex script
rendering issues in OpenType.
Exactly this is the problem. Direct Unicode->Glyph mapping is *soo*
easy...
Werner
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:43 EDT