> I would propose adding (to draft-15): > > 6. Redundant script tags which are the same as the suppressed > script for a language are not present. (technical hat) We can't do that. There are valid reasons why someone might use a script subtag in this way (we actually spell some of them out). Canonicalization that removes real subtag distinctions would be bad, even if we suggest that such a tag choice is not wise in the vast preponderance of cases. > > In addition, I would like to ask about fonipa and fonupa. Currently > in 4645 they have no prefix, which makes sense since they can be > used with any language. But they can only be used with Latn script. > In fact, they imply Latn script. So what should the canonical form > of Eastern Bru in IPA be? > > bru-Latn-fonipa > > or can we in some way indicate the implication of Latn by fonipa > and just use: > > bru-fonipa Not in the registry, but not everything can or should be done in the registry. If you use 'fonipa' you most certainly understand that the text is in some inscrutable phonetic alphabet. A script subtag is probably not a good tagging choice at all in this case. > > I would suggest that if we want to do this, then specifying a > prefix in the record for fonipa, of Latn would be sufficient > although we may want some wording somewhere along the lines of: > > Where the prefix tag for a variant consists of a script, that > variant implies the presence of that prefix and the prefix is > required not to exist in the canonical form of the tag. Prefixes never consist of "a script". They consist of language ranges. We could achieve the effect you want by changing the range to an extended language range. The record would then look something like: Subtag: fonipa Prefix: *-Latn But this would have the negative effect of disallowing a tag such as "bru-fonipa", which I think is a perfectly fine tag: it would require the form "*-Latn-*-fonipa". Furthermore, 'fonipa' doesn't really imply "Latin script". It implies characters from the latin script, certainly. But this is not the same thing. Especially since some of the IPA characters aren't Latin script at all! Best Regards, Addison _______________________________________________ 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.