From: Peter Kirk (peterkirk@qaya.org)
Date: Mon May 10 2004 - 09:34:43 CDT
On 07/05/2004 15:59, Michael Everson wrote:
> At 17:10 -0400 2004-05-07, jcowan@reutershealth.com wrote:
>
>> This would only be the *default* rules. Unicode-savvy sort programs can
>> accept "tailorings" that make the rules different, like the Swedish
>> tailoring
>> that makes a-ring, a-umlaut, and o-umlaut sort after z instead of in
>> their
>> default places with a and o.
>
>
> As I said, they would be the *tailored* rules. Mixing scripts would go
> against the current practice of ISO/IEC 14651.
Well, we are not talking about ISO/IEC 14651 but about Unicode. Is there
any really good reason not to mix two scripts, which are according to
many people actually variants of one script but which are (if your
proposal is accepted) seperately encoded for the convenience of some
scholars? This sounds to me like the kind of rule which is made to be
broken. If all the 22 CSWA scripts are collated together by default,
this would significantly reduce the objections to encoding them as
separate scripts. We can perhaps consider them as a family of congruent
scripts. Of course we might then think that there are other such
families, e.g. the different Indic scripts, but how to collate them
should depend on Indian etc custom.
-- Peter Kirk peter@qaya.org (personal) peterkirk@qaya.org (work) http://www.qaya.org/
This archive was generated by hypermail 2.1.5 : Mon May 10 2004 - 23:59:05 CDT