RE: Latin w/ diacritics (was Re: benefits of unicode)

[Marco Cimarosti]

Peter Constable wrote:
> Why would you encode presentation form glyphs in the PUA if you don't
> expect them to be encoded directly in documents. "Smart font" 
> rendering systems map character codes into glyph ids, and so these 
> glyphs don't need to be encoded in the cmap.

I may be wrong, but my impression is that it is considerably easier to act
on characters than on glyph ids inside fonts.

Imagine that your language needs a + dieresis + tilde + acute, and no smart
font around supports this. But you desperately need to write in your
language *today*, and you want to do it with, say, Word.

If you are a smart user who can use a font editor and write Word macros, you
could add a precomposed glyph in the PUA (copying the glyph for "ä" and
adding pieces of glyphs for "ñ" and "é"). Then you build a pair of macros:
the first one runs after file loading and substitutes the composite sequence
with the PUA character; the second macro runs just before saving and
substitutes the PUA character with the proper sequence.

I would definitely agree to call this a dirty hack... But would it be so
evil? Consider that, as soon as a new version of your favorite OS delivers
full support for your special sequence, you can throw away all your hacks:
the important thing is that the electronic corpus that you have produced is
still readable.

_ Marco