🚀 go-pugleaf

In article <1ii6bkINNf6c@rodan.UU.NET> of comp.std.internat,
  avg@rodan.UU.NET (Vadim Antonov) writes:
> 
> >A pure locale-system clearly won't do in multilingual environments.
> >Nonetheless some things are, IMHO, best handled with locales.
> >Perhaps it should be possible to specify the default language (like,
> >in an environment variable) separately for each script one is
> >concerned with (setenv LANG 'LATIN:finnish;CYRILLIC:russian;HAN:chinese')
> >or whatever) and fall to a default in the rest.
> 
> It is not enough. Ukrain, say, use characters of both Cyrillic and
> Latin scripts, etc.
> 

I'd suggest some form of multiple locales, i.e. 
setenv LANG ukranian-swedish-ipa-greek-farsi-japanese

In this model, each of the sorting algorithms would assign sorting
values (integers, that may or may not be identical) to each character
recognized in that language.  Any character undefined (e.g. KANJI HITO
for Ukranian) is assigned MAXINT and thus is sorted last.  The sort
proceeds with the first language as the primary key, second language
as the secondary etc.  That means not only that the text in the
different languages will show up in approximately the order listed,
but also that multiscript languages can be easily accommodated.  There
are languages that use characters from at least three scripts, so the
need is there.

	/hpa

-- 
INTERNET:  hpa@nwu.edu   TALK:      hpa@delta.eecs.nwu.edu
BITNET:    HPA@NUACC     IBMNET:    16331@IBMX400
HAM RADIO: N9ITP         NeXTMAIL:  hpa@lenny.acns.nwu.edu
while ( 1 ) ; cp /dev/zero /dev/null & end