Script codes in RFC 3066

[Caoimhin O Donnaile]

On Wed, 9 Apr 2003, Michael Everson wrote:

> I don't see how current practice has much to do with your (very
> interesting) database idea though.

If "language hetrarchy aware" software is going to be the norm in
the near future it could have the following implications:

 - There should be more of a drive to label web pages correctly -
   e.g. as "Pfälzisch" rather than as "German"; or as "Scots"
   rather than as "English" if this is the case - because users will
   (hopefully!) set their browsers and searches to accept the
   appropriate wider category and still find them.

 - More pages and other resources should be labelled with the
   best available language information, e.g. "Frisian", rather
   than being left unlabelled because the exact language is
   unknown.

 - Even more incentive to vastly expand the set of codes allowed
   in information processing in the near future.  Less fear of
   "getting it wrong" since the database will retain basic
   information on obsoleted codes.

 - More incentive to tag at the "sub-language" level, or at least
   be liberal in the definition of "language".

 - Little incentive to use hyphens in new language codes, since
   the database would provide much better hierarchic information.
   Existing codes with hyphens could be treated as atomic strings,
   with the hyphens purely for human ease of reading.  Alternatively,
   there would be the possibility of evolving towards an entirely
   fixed-length system of language codes for neatness, if this
   were felt to be desirable, with the previous hyphenated codes
   retained in the database as obsoleted codes.

Caoimhín