Re: Proposed new characters updated in Pipeline Table

2011/8/15 Petr Tomasek <tomasek_at_etf.cuni.cz>:
> On Mon, Aug 15, 2011 at 07:38:07PM +0200, Philippe Verdy wrote:
>> 2011/8/13 mmarx <mmarx_at_zedat.fu-berlin.de>:
>> >>> I attach a Garshuni document from
>> >>> Beit al-Qammar showing Arab vowel
>> >>> marks -- just as the Syriac communmties
>> >>> are using Syriac vowel marks in Arabic
>> >>> script text -- and (in the second line
>> >>> on the left) wasla above olaph.
>> >>>
>> >>> So whatever the status of wasla mark
>> >>> will be fore Arabic, for Syriac/Garshuni
>> >>> it is need as a normal mark in normal text.
>> >
>> >> Unicode cannot encode a combining Wasla (because of various stability
>> >> policies), so if Syriac needs a Wasla to be shown only over a letter
>> >> or two, one needs to propose precomposed characters for them. Just
>> >> like the existing Arabic Alef-Wasla.
>>
>> Why not? If the character is new, it can perfectly be encoded with
>> whatever character property is needed, including with a non-zero
>> combining class, if it fits. The stability policy aboud combining
>
> But as soon as you have a combining wasla in the Unicode you would
> have to change the decomposition mappings for Alef-Walsa (because
> from that point on you would have two possibilities how to encode
> this combination).

No. The decomposition mapping of Alef-Wasla would not be changed at
all, even if this would create two "confusable" encodings for the
Alef-Wasla combination, they would remain different under canonical
equivalence and compatibility equivalence.

The stability of mappings and decomposition remains a rule, and Wasla
can still be encoded separately, if there are good reason to encode it
in the Arabi script other than its current visual occurence in
Alef-Walsa, which will remain special and unchanged. Instead of
decomposition mappings, some kind of approximate unification can still
be setup by collation (with minor differences only at the code point
level).

> However, If a special "Syriac Wasla" was encoded as a combining
> character than no problems would arise for the Arabic script.

This is an independant issue, this does not influence the encoding of
Wasla in the Arabic script.

But it's also possible, if the separate Wasla is needed in both Arabic
and Syriac, to unify them in a single encoding, and then use a Syriac
Wasla in Arabic (just like its's possible to use any other Arabic
combining characters with Syriac base letters, or the reverse).

Philippe.
Received on Mon Aug 15 2011 - 14:15:25 CDT