Review period; Nepali and Oriya

[Doug Ewell]

Gordon P. Hemsley wrote:

> Ah, sorry, forgot to include the link:
> http://unicode.org/repos/cldr/trunk/common/supplemental/supplementalMetadata.xml
>
> Every <languageAlias reason="macrolanguage" /> maps a microlanguage
> code (@type) to a macrolanguage code (@replacement). This seems to be
> for the same reason as the existence of extlangs, except that this
> eliminates the ability to use macrolanguage codes for their
> macrolanguage usage.

OK, I see the data now. No, we would not want to do that in BCP 47.

> For example, 'cmn' cannot be used as Mandarin; instead, 'cmn' is
> mapped to 'zh', meaning that "Chinese" usually really means
> "Mandarin", so that's now what 'zh' means.
>
> But it does have its uses; for example, it's pretty handy if
> you have legacy locales like 'zh'—they are assumed to be Mandarin
> anyway, so no need to switch to 'cmn'.

This works well or not so well, depending on the meaning of "usually" 
and on whether one happens to be dealing with the "usually" case. 
Locations and circumstances exist in which "Chinese" definitely does not 
mean "Mandarin."

CLDR contains a good deal of "best guess" data like this (e.g. 
http://unicode.org/repos/cldr-tmp/trunk/diff/supplemental/likely_subtags.html) 
which is probably useful for many applications, but which is better 
handled in BCP 47 by employing matching strategies such as those 
suggested in RFC 4647.

--
Doug Ewell | Thornton, Colorado, USA
http://www.ewellic.org | @DougEwell ­