From: Richard Wordingham (richard.wordingham@ntlworld.com)
Date: Sat Mar 25 2006 - 08:06:36 CST
Kenneth Whistler wrote:
Subject: Re: AA versus TALL AA (Was: the Myanmar thread from h e double
hockeysticks)
Is this a Unicore escape?
>> Scenario 1
>>
>> This results from Everson's advocated solution, disunifying
>> based on glyph shape.
>>
>> Encoding Rendering rule Reading rule
>>
>> Burmese: xa yA a --> {a} {a} --> a
>> a --> {A} {A} --> A
>>
>> S'gaw Karen: xA yA a --> {A} {A} --> A
>
> Sorry, to beat you all to the punch, recte:
>
> Scenario 1
>
> This results from Everson's advocated solution, disunifying
> based on glyph shape.
>
> Encoding Rendering rule Reading rule
>
> Burmese: xa yA a --> {a} {a} --> a
> A --> {A} {A} --> A
>
> S'gaw Karen: xA yA A --> {A} {A} --> A
Aren't the reading rules more like '{a} --> A and {A} --> A'?
For Burmese, isn't this akin to using undotted 'i' (U+0131) in the English
word 'fish' because good typography leaves it ligated and undotted? The
motivation for the tall form in Burmese seems to be to distinguish consonant
plus vowel from another consonant. Thus, under Scenario 1, one would
encounter:
Burmese: Encoding: wa Rendering: wa Reading: t@ ...aa wa! (or
something similar!) as opposed to the current Unicode:
Encoding: wa Rendering: wA Reading: wa
One question that was not considered worthy of consideration in the proposal
( http://std.dkuug.dk/jtc1/sc2/wg2/docs/n3043.pdf ) was how accurately the
Burmese actually select the correct form when typing (or indeed, when
writing by hand). I guess that one reason for Sgaw Karen dropping the round
form was that it was simpler to teach a single glyph shape.
Moving slightly eastwards, the decision here sets a precedent for the Lanna
script. My Lanna text book seems quite sloppy in its choice. The practical
rules may also be quite complex, though I stopped worrying when I was
advised that a solution using variation selectors could leave the glyph
choice in their absence to the application. Medial RA and subscripts with
ascenders on the preceding consonant seem to inhibit tall AA, but it seems
very much to be a matter of style and aesthetics in the Lanna script.
I suppose the best precedent for having two characters is the distinction
between U+017F (Latin small letter long s) and U+0073 (Latin small letter
s). However, I recall that distinction being justified by the selection
being neither easily algorithmically determinable nor stable over time.
Richard.
This archive was generated by hypermail 2.1.5 : Sat Mar 25 2006 - 08:10:54 CST