I wonder if it would be acceptable to punt MultiLingual messages to
RichText?
I don't know. That's a very difficult question. Until we started
this discussion, we basically only had:
Content-Type: text/subtype; charset=foo
MIME said that the subtype "plain" was for "unformatted" text. What,
exactly, does this mean? John has been suggesting that perhaps
language tagging could be considered to be "markup", and therefore not
part of "text/plain". Is he right?
We have been talking about various concepts:
procedural markup (e.g. bold, italic)
descriptive markup (e.g. emphasis, alternative vocabulary)
language tagging (e.g. French, Japanese)
font tagging (e.g. Helvetica, Ryumin Light)
characters (e.g. 'a')
glyphs (e.g. the various glyphs for 'a')
Now, the problem seems to be that Masataka wants to distinguish the
Han things. Whether we call those things "characters" or "glyphs"
probably depends on one's point of view. So there's a difficulty
there.
Another difficulty is "Where should we put the info in the headers?"
Do we need another parameter such as "glyphcode=foo"? Do we want to
put languages in the charset parameter, in a separate parameter, in a
separate header, or in a text subtype? Etc etc etc.
Interesting debates!
Erik