RE: [Ltru] Re: Macrolanguages, countries & orthographies

Mark wrote:

1. Your personal opinion contradicts what I heard as a rough consensus
earlier, that fr is not a superset of fro or frm. The highest priority is
that the interpretation be well-defined and consistent; I think your opinion
would be reasonable, but it doesn't seem to be the consensus. 

Thank you.  May be we should work on making it consensus.

2. If you asked me right now up-or-down on ISO 649-6, I'd say absolutely
not, since (a) it would introduce all kinds of duplicate encodings, 

I believe we worked out a system of flagging the ISO 639-6 data in order
that duplicate encodings would not be introduced within the Registry; a
simple procedure which is currently being actioned by GeoLang (compilers of
the ISO 639-6 data).

and (b) there has been no clear rationale given that the other information
is worth adding. 

Hmmm... I thought we were having a discussion on variants and historical
variants, something that is facilitated by the proposed ISO 639-6 code.

[snip]But as per earlier messages, end users have very little idea of
language taxonomies. A far better interface for narrowing choices would be
one that lets them pick a country, and then languages in use in the country.
But all of that is beside the point -- UI is *not* the goal of BCP 47.

But if one can have both... a faceted approach... what a winner :-)  BTW
FYI, ISO 639-6 data is also linked to ISO 3166-1 and ISO 15924.  I believe
we have had the discussion wrt numbers of language selections in some
countries but I can certainly see that in some cases this would be
preferable.  Tell me, hypothetical scenario, what would you recommend a user
to do when their language is not listed under their country of residence as
there are only two speakers of Khanty residing in the UK?  Just a thought.  

The biggest problem on the score of pick lists is that language names and
country names in the users language are required; linked to the ISO
639/3166-1 code. 

Best 

Debbie