Re: [Ltru] Does 'de' really mean "only standard German"?

["Mark Davis" <[e-mail]>]

ISO 639-1 and ICU 639-2 *didn't* do this in the RFC 4646 time frame, and so
RFC 4646 doesn't either.

ISO 639-3 chose to interpret 'de'/'deu' as "Standard German", and 'ar'/'arb'
as "Any Arabic", but someone reading 639-1/2 and applying RFC 4646 *in good
faith and with the information available in the standard* would have no idea
that there was any fundamental difference in the way that they were going to
be defined in 639-3.

ISO 639-3 does have much more concrete definitions (all to the good), but
someone implementing ISO 639-1/2 or RFC 4646 or any of its predecessors
would have no idea that German would get defined with one model, and Arabic
get defined with a completely different model. It was reasonable for someone
implementing RFCs 1766, 3066, and 4646 to interpret "de" as Standard German
(Hochdeutsch) and "zh" as Standard Chinese (Mandarin). It would *also* have
been reasonable for someone to interpret 'de' as Any German (and up until
2006-03-08, encompassing Swiss German), and 'zh' as Any Chinese (thus
encompassing Cantonese).

The standard was underspecified.

I am not arguing against the way ISO 639-3 has done things; that is water
under the bridge -- but we need to recognize that because of the imprecision
of the previous ISO 639 standards, we can't make reasonable interpretations
of RFCs 1766, 3066, and 4646, widely followed in industry, be
non-conformant.

Mark

On Tue, May 27, 2008 at 12:26 PM, John Cowan <[e-mail]> wrote:

> Randy Presuhn scripsit:
>
> > I'm very surprised by this claim.  I thought 'de' encompassed any
> > variety of German that wasn't covered by some other tag (like gsw),
>
> It does encompass national varieties as well as different orthographic
> reforms of Standard German.  But it does not cover the various other
> West Germanic languages (Mundarten), High or Low, spoken in Germany.
>
> How do we know this?  The ISO 639-2/639-3 equivalent of 'de' is 'deu',
> and when we look at http://www.sil.org/iso639-3/documentation.asp?id=deu
> (which defines 'deu'), we are pointed to encyclopedic information at
> http://www.ethnologue.com/show_language.asp?code=deu , which makes it
> clear that the denotation of 'de' is Standard German, the language
> spoken by almost a million people in Kazakhstan.
>
> > just as 'fr' encompasses many flavors of French.
>
> For whatever reasons, Ethnologue (and therefore ISO 639-3) does not
> distinguish the various langues d'oil except for Picard and Walloon,
> so all the rest are indeed lumped under 'fr'.
>
> --
> A rose by any other name                            John Cowan
> may smell as sweet,
> http://www.ccil.org/~cowan <http://www.ccil.org/%7Ecowan>
> but if you called it an onion                       [e-mail]
> you'd get cooks very confused.          --RMS
> _______________________________________________
> Ltru mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/ltru
>



-- 
Mark

_______________________________________________
Ltru mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/ltru