Dana S Emery writes:
We can do that either by defining more headers (as Danny
suggests), or by allowing parameters:
Content-language: en; dialect=us; accent=bronx;
hostility=high
or whatever. We don't have to do it all the first time
around, so long as we leave room to build on whatever we
do.
thank you steve, very well put.
Agreed. However, I think that I would like to see someone (some sort
of linguist, just as an example) tell us that 'yes, the data allowed in
the header(s) you have defined will enable a large group of people to
specify their language of choice' or, have someone say 'wait, you are
exluding this large group of people, why can't you ask for so-and-so
information instead'. I don't think that Dana and I are in disagreement
on this, hence Dana's research over the past few days.
I guess my fear is that a Content-language will be defined that isn't
very usuable, or not very extensible without throwing away past work
in order to fix any wholes. Thats all. At the moment I am fairly
convinced that it is best to go with 1 "Content-language" header and
to make that fairly bulky.
BTW: I am not sure how my comments are being taken so just want to
restate the fact that I am 100% for the addition of a language
content in MIME.
--------------------------------------------------------------------------------
Neophytos Iacovou
Honeywell Systems & Research Center
iacovou(_at_)src(_dot_)honeywell(_dot_)com
Minneapolis, Mn, 55418