Vedic Accents and Unicode

[Paul G. Hackett]

At 5:55 PM +0100 3/23/06, Ferenc Ruzsa wrote:
>Yes, combining diacritics are a possible solution, but it is much 
>nicer and more stable (and easier to convert) when we have a single 
>character representing one phoneme (letter).

This may be true, but such an approach would not conform to the 
principles of Unicode, since the standard aims at encoding characters 
(abstract linguistic components) and not glyphs (surface / appearance 
forms).  See:

http://www.unicode.org/reports/tr17/#CharactersVsGlyphs

FYI, early versions of the standard did encode some composite forms, 
but these have since been deprecated.

Paul Hackett
Columbia University