[Ltru] To invert or not to invert?

["Doug Ewell" <dewell at adelphia.net>]

The new RFC 4646bis draft does finally touch on the issue of mapping the 
language names from ISO 639-3 to Description fields in the Registry. 
But there is still some wiggle room.  From Section 3.1.4:

  Each 'Description' field MUST be unique, both within the record in
  which it appears and for the collection of records of the same type.
  Moreover, formatting variations of the same description MUST NOT
  occur in that specific record or in any other record of the same
  type.  For example, while the ISO 639-1 code 'fy' contains both the
  descriptions "Western Frisian" and "Frisian, Western", only one of
  these descriptions appears in the registry.

  ...

  When creating a new registry entry, duplicate, redundant,
  conflicting, or otherwise problematic descriptions MUST either be
  corrected or omitted.  Parenthetical comments, inverted names, and
  other irregularities SHOULD be regularized according to the
  guidelines used to update the registry in [registry-update].

The first passage indicates that EITHER the inverted form of a name OR 
the uninverted form must appear as a Description.  The second passage 
speaks of "regularizing" inverted names, which suggests that it is the 
uninverted name which is used, but does not say so explicitly, and it 
essentially pushes the problem to RFC 4645bis, which is where it belongs 
as far as the bulk update is concerned.

Either the inverted or uninverted name would equally serve the purpose 
of matching the Description fields for a language subtag to the ISO 
standard,  I haven't seen a requirement from anyone that *all* of the 
various ISO names, inverted and uninverted, must be matched.

What is the preference of the WG on this issue?  Should the Registry use 
"Western Frisian" or "Frisian, Western", given that it must use one and 
only one?  I can do it either way with roughly equal ease.  Please send 
your preference to the list.  I'd like to do a 4645bis draft this 
weekend.  If I don't hear a strong preference for "inverted," then I'll 
go with "uninverted" since that seems to be the direction in which 
4646bis leans.

--
Doug Ewell  *  Fullerton, California, USA  *  RFC 4645  *  UTN #14
http://users.adelphia.net/~dewell/
http://www1.ietf.org/html.charters/ltru-charter.html
http://www.alvestrand.no/mailman/listinfo/ietf-languages


_______________________________________________
Ltru mailing list
Ltru at ietf.org
https://www1.ietf.org/mailman/listinfo/ltru