John Hudson wrote:
That's correct: General purpose language information does not belong into the plain text data stream. John Cowan replied:> I don't think anyone is questioning that language tagging is a good and > useful thing. The question is whether using character codepoints as > language identifiers is a good thing. I'm inclined to the view that it is > not, and that language tagging should be handled, along with most (all?) > other tagging, at a higher level.
That's all fine and dandy - but unless there's demonstrable implementation of this technique anywhere the conclusion is that it's a solution in search of a problem and as such subject to cleanup. [Since we can't remove them, we wouldI think it's time to remember the limited purpose for which Plane 14 tagging was created: plain-text protocol messages. The idea is that when contacting an IETF-protocol server, it should be able to report back in various languages, using plain-text tagging to indicate which language you are getting (or, if it reports in multiple languages, which is which).This was considered to be a situation where heavyweight (XML, etc.) metadata was unnecessary: --> RETR 32 <-- 522 LTAG{en}I have no clueLTAG{art-lojban}mi na jimpe
deprecate them, so that countless legions of implementers can forget worrying
about a feature deemed desirable but never put into practice.]
If that premise (neat idea but noone does it) is disproven then the status
quo ante should remain -- limited use for plaintext protocols only.
I've seen lots of discussion about the purpose/potential of the tags - much of it misguided - but, unless I missed it in the torrent, there seems to be no smoking gun of IETF style implementations, many years after this solution was demanded for them.
Case closed.
A./
