ISO 639 JAC decision re mo/mol
randy_presuhn at mindspring.com
Tue Nov 4 19:55:39 CET 2008
> From: "John Cowan" <cowan at ccil.org>
> To: "Doug Ewell" <doug at ewellic.org>
> Cc: <ietf-languages at iana.org>
> Sent: Tuesday, November 04, 2008 7:30 AM
> Subject: Re: ISO 639 JAC decision re mo/mol
> So we are dealing with a total of 22.26 million speakers of
> Romanian/Moldovan, of which a full 99% are using Latin script in Romania
> and Moldova, versus only 1% using Cyrillic script in Transnistria.
> I call that overwhelming.
> > Thus: the overall language that will be represented by 'ro', in its
> > various flavors, is not overwhelmingly written in any one script, and
> > consequently the existing Suppress-Script for 'ro' will no longer be
> > appropriate and should be removed.
> Au contraire.
Whether these back-of-the-envelope numbers are correct or not, this
kind of argument-by-the-numbers illustrates why RFC 4646 does NOT
give an absolute numeric or relative percentage threshold for "overwhelming."
The original motivation for Suppress-script was backwards compatibility
with existing tagging practice. One might consider it an attempt to weakly
enforce "tag wisely" practice, motivated by a desire to avoid script subtags
in situations where they would be of negligible value.
The question here is whether tagging or tag matching / filtering would work
better or worse for these user communities with or without the Suppress-script.
More information about the Ietf-languages