Re: Characters and glyphs - was Re: [I18n]bdftruncate again ;)

[Aidan Kehoe <kehoea at parhasard dot net>]

 Ar an 24ú lá de mí 1, scríobh Dr Andrew C Aitchison :

 > A simpler example is that 'A' and 'a' are the same character, but 
 > different glyphs. If Unicode was being even-handed, it would
 > include one but not both of these glyphs, and application software
 > would have to calculate which glyph to use each time the character
 > was drawn.

So, for example, it would examine the semantics of a dissertation to
see whether `God,' was being used as a proper or common noun? Hmm. And
know that `etc.' is an abbreviation, and thus one shouldn't capitalize
the following word, unless it actually *does* end a sentence, which is
something you can't actually tell without knowing, again, semantics?

Upper-case vs. lower-case in Roman script is much more involved
problem than Brian Stell's 

	"glyph for the first letter in a word
	glyph for a letter in the middle of a word
	glyph for a letter at the end of a word
	glyph for a letter that is a single letter word"

for Arabic, and it wouldn't have been (or be) constructive to force
application software to incorporate such heuristics.

 > This example shows how biased Unicode is towards Latin characters.

-- 
`... when the elephant man broke strong men's necks, when he'd had too 
many Powers, ...'