- From: Carrasco Benitez Manuel <manuel.carrasco@emea.eudra.org>
- Date: Wed, 21 Oct 1998 15:12:16 +0100
- To: "'Harald Tveit Alvestrand'" <Harald.Alvestrand@maxware.no>, www-international@w3.org, tc46sc2@elot.gr
>At 12:15 20.10.98 +0100, Carrasco Benitez Manuel wrote: >>I could write a new proposal using a (new) "t" in the primary >>language tag. >> >this sounds very wrong to me; t isn't a language, nor is t- a family >of languages that shares something in common. In RFC 1766, the first characters does not seems to denote a language or family of languages. Namely i- for IANA x- for experimental Again, the syntax is secondary and this could be indicated in another way. >I would prefer, if this is going to be done at all: > >el-script-latin (Greek written with Latin letters) >el-script-latin-farouk (Greek written with Latin letters according to > Farouk's set of transliteration rules) >en-cockney-script-ipa (Cockney English written in the International > Phonetic Alphabet) >fr-CA-script-braille (French-Canadian written in Braille) > >Note that there are many examples where one would put the "-script-" tag in >positions other than the first one; Canadian French is still fr-CA no >matter what the representation is. Fine. Perhaps using another term; "script" is short and descriptive, but it could leave out some transformations. The name of the transformation could carry more information to avoid having a syntax that is too complex. e.g. el-tran-latin (Greek transformed using "latin" ) el-tran-farouk (Greek transformed using "farouk" ) or el-xxx-latin (whatever string is choosen) el-xxx-latin t-el-latin (assuming t- syntax) t-el-farouk The value of the term "latin" and "farouk" would have to be define in in the register. For example, in the case of "latin" could refer to: - Latin script - Latin, *the* language - Generic for Latin languages Additional rules would be attached also. >I would also prefer the script to be documented as a convention, and >have each specific new scheme or usage registered as a language tag >in the language tag registry. It never hurts to register.... Totally agree. There must be a register with the nomenclature, the "authority" for each transformation, implies, etc. I contacted the TC46/SC2. Regards Tomas
Received on Wednesday, 21 October 1998 10:13:10 UTC