Variant tags for sl-rozaj: History and preliminaries Draft of message

han.steenwijk at han.steenwijk at
Fri Jun 15 08:45:20 CEST 2007

Earlier on, I enthusiastically said that "even the order of the variant
tags within the language tag is determined by the Prefix fields."

However, the following statement from RFC 4646 makes me somewhat uncertain
on this point:

An implementation that claims to be validating MUST:
For variant and extended language subtags, if the registry contains one or
more 'Prefix' fields for that subtag, check that the tag matches at least
one prefix. The tag matches if all the subtags in the 'Prefix' also appear
in the tag. For example, the prefix "es-CO" matches the tag
"es-Latn-CO-x-private" because both the 'es' language subtag and 'CO'
region subtag appear in the tag.
(From: 2.2.9.  Classes of Conformance)

This makes me think that, with a prefix 'sl-rozaj' for the variant subtag
'biske', both "sl-rozaj-biske" and "sl-biske-rozaj" would be valid tags,
as in either case the 'sl' language subtag and 'rozaj' variant subtag
appear in the language tag. Has it somewhere been determined that the
subtags that make up a prefix should all occur to the left of the variant
subtag to which the prefix applies? Actually, that is what the notion
'prefix' entails.

If no such provision has been made, could the Preferred-Value field be
used to order variant subtags?


Prof. Han Steenwijk
Cattedra di Lingua e Letteratura Slovena
Universita' di Padova
Dipartimento di Lingue e Letterature Anglo-Germaniche e Slave
Sezione di Slavistica
Via Beldomandi, 1
I-35139 Padova

tel. 049 8278669
fax. 049 8278679

More information about the Ietf-languages mailing list