As far as I can see, UNICODE requires one piece of profiling
information: Its encoding method.
According to Otha-san, it would require one more sentence, saying
that "the algorithm described by ... and ... in <the Unicode book>
is followed to determine the context in which to render the text.
This may not always give the result intended by the sender when
the recipient and sender do not have the same context"
No more than this needs to be said, I think. And I think no more *can*
be said, until the standards are changed.
But why are we taking this to the 822 list, and not to -charsets?