In message <9903020955.AA22614@unicode.org>
Kevin Bracey <kbracey@e-14.com> wrote:
> 11AC (NJ) ~= 11AB (N) + 11BD (J)
>
> Are these really only compatibility decompositions, not canonical
> decompositions? Is G A NJ (1100 1161 11AC = AC05) really not canonically
> equivalent to G A N J (1100 1161 11AB 11BD)? This would imply that the
> composed form NJ has extra formatting information that is lost by
> decomposing (conformance requirement D20). What formatting information?
>
> In section 3.10, stage 1 of Hangul Syllable Composition says to "compose
> the conjoining jamo wherever passible". As the only compositions are
> compatibility compositions, does that mean that the Hangul syllables are
> really only compatibility compositions of the conjoining Jamo?
>
Okay, I've found what TR8 has to say, and it's clarified it a bit. Hangul
syllable composition is definitely a canonical equivalence, but it looks
like we're pretty adamant that NJ is only compatibility equivalent to N+J.
I still don't fully understand the logic here. Why is G+A+NJ different
to G+A+N+J? The syllable still forms in the same way, and I can't see how/
why it would be rendered differently.
Is it just to simplify processing for systems that assume a syllable
consists of 1 L + 1 V + 1 T, rather than an arbitrary number of each, as
Unicode allows?
-- Kevin Bracey, Senior Software Engineer Acorn Computers Ltd Tel: +44 (0) 1223 725228 Acorn House, 645 Newmarket Road Fax: +44 (0) 1223 725328 Cambridge, CB5 8PB, United Kingdom WWW: http://www.acorn.co.uk/
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:44 EDT