The question is whether xx and yyy can have separate S-S values. I think it is *possible* that this could happen, it just isn't apparent that we have the problem at present. For example, someone pointed out to me that some forms of Arabic use(d) Latin script---it is just that these forms are now extinct.
I don't think that xx-yyy and yyy can have different S-S values for the reason John points out: a language either is or is not written in "just one" script. Other combinations seem plausible though:
Xx yyy
SS a a
SS a b
SS - a
SS a -
SS - -
Where (a) and (b) are different scripts and (-) represents no SS present. The most unlikely cases are the second and fourth ones (xx has a script and yyy does not or has a separate script). In these cases, a language could be "overwhelmingly written" in a script while a fairly small, usually unremarked, encompassed language is the exception.
Addison
Addison Phillips
Globalization Architect -- Lab126
Internationalization is not a feature.
It is an architecture.
> -----Original Message-----
> From: ltru-bounces at ietf.org [mailto:ltru-bounces at ietf.org] On
> Behalf Of John Cowan
> Sent: Monday, July 21, 2008 10:05 PM
> To: Randy Presuhn
> Cc: LTRU Working Group
> Subject: Re: [Ltru] Changes to draft-4645bis resulting from new
> draft-4646bis
>
> Randy Presuhn scripsit:
>
> > > Problem: If you have some flavor of Arabic, denoted by an
> extlang, that
> > > isn't written overwhelmingly in the Arabic script, how do you
> interpret
> > > the absence of a script subtag? (I personally don't think this
> is a
> > > huge real-world problem, but it's the sort of thing we in LTRU
> love to
> > > dwell on.)
> >
> > For precisely this reason, the Supress-Script (if any) associated
> with
> > a particular subtag should not be presumed to indicate what the
> > Supress-Script would be for something else, even in the case of a
> > primary-extlang combination, even if they are overwhelmingly
> likely
> > to have the same value. I don't think it is reasonable to
> *require*
> > every extlang to have the same suppress-script as the
> corresponding
> > primary language tag.
>
> -1
>
> Having a Suppress-Script or not, and what particular value it has,
> is
> a property of a language, not a way of describing that language.
> It is
> not reasonable for xx-yyy to have a different S-S value from yyy,
> since
> xx-yyy and yyy are synonymous, any more than it would be reasonable
> for
> he and iw to have different S-S values. It *is* reasonable for xx-
> yyy
> and xx to have different S-S values.
>
> > (Particularly if we remember that the original intention of
> > Supress-Script was compatibility with existing data that lacked
> script
> > subtags, and would not be an issue for newly regstered languages
> since,
> > by definition, there would be no legacy data.)
>
> That's true. Unfortunately, several xx-yyy forms are legacy.
>
> > Putting in a rule to require them to "propagate" or to be the
> same
> > would be a false economy, in my opinion.
>
> How a false economy? It would be an economy (true or false) not to
> provide S-S headers for both the extlang and the corresponding
> language
> tag. It would be inconsistent to allow them to have different
> values.
> IMHO it is less confusing to provide the header in both places.
>
> --
> John Cowan cowan at ccil.org http://ccil.org/~cowan
> Female celebrity stalker, on a hot morning in Cairo:
> "Imagine, Colonel Lawrence, ninety-two already!"
> El Auruns's reply: "Many happy returns of the day!"
> _______________________________________________
> Ltru mailing list
> Ltru at ietf.org
> https://www.ietf.org/mailman/listinfo/ltru
_______________________________________________
Ltru mailing list
Ltru at ietf.org
https://www.ietf.org/mailman/listinfo/ltru
Note Well: Messages sent to this mailing list are the opinions of the senders and do not imply endorsement by the IETF.