> From: Phillips, Addison [mailto:addison at amazon.com] > Sent: Tuesday, July 08, 2008 9:58 PM > This section, as noted, is a mess. I've altered it significantly... A further problem I note is that the portion starting with "A language tag is in canonical form when..." is inconsistent: parts of it describe conditions satisfied by canonical form (as expected by that into sentence), and parts describe instead what changes to make to map a tag to a canonical equivalent. Something like the following seems like it might be better: --------------- A language tag is in canonical form when: 1. The tag is well-formed according the rules in Section 2.1 (Syntax) and Section 2.2 (Language Subtag Sources and Interpretation). 2. The tag is not a redundant or grandfathered tag that has a Preferred-Value mapping in the IANA registry. 3. The tag does not contain language, script, region or variant subtags that have a Preferred-Value mapping in the IANA registry. 4. Any extensions are ordered into case-insensitive ASCII order by singleton subtag. For a given tag, an equivalent tag in canonical form can be derived by applying the following changes: 1. Redundant or grandfathered tags that have a Preferred-Value mapping in the IANA registry MUST be replaced with their mapped value. 2. Subtags of type 'extlang' SHOULD be mapped to their Preferred-Value. 3. Other subtags that have a Preferred-Value field in the IANA registry MUST be replaced with their mapped value. 4. If more than one extension subtag sequence exists, the extension sequences MUST be re-ordered into case-insensitive ASCII order by singleton subtag. --------------- I've left out some of the substantive wording that is in the current draft (e.g. explaining the origin of the grandfathered and reduandant tags); if this suggestion is accepted in principle, you might want to figure out appropriate ways to work that content back in. There's another concern raised by the changes you made in that the canonical derivation for tags is vague wrt extlang. I suggest we leave that for a separate thread. Peter _______________________________________________ Ltru mailing list Ltru at ietf.org https://www.ietf.org/mailman/listinfo/ltru
Note Well: Messages sent to this mailing list are the opinions of the senders and do not imply endorsement by the IETF.