Re: [tlug] Japanese Encoding - which one?

[Tobias Diedrich <ranma@example.com>]

Stephen J. Turnbull wrote:

>     Lyle> Content-Type: text/plain; charset="us-ascii"
> 
> Why does this happen?  Because it doesn't matter whether your
> conversation partner speaks US ASCII, ISO 8859-1, Shift JIS, EUC-JP,
> UTF-8, or ISO-2022-JP, _every one_ encodes that message with the same
> bytes.
> 
> Now, suppose you send the message with "ISO-2022-JP" as the content
> type; then dumb clients will say "I don't know how to do that; save to
> a file?"  On the other side, even a dumb client able to handle any of
> the others can treat "US-ASCII" as an alias for whatever it _does_
> handle.
> 
> I'm pretty sure the standard for Content-Type recommends that this
> kind of weakening of requirements (especially Unicode -> ISO-8859-1 ->
> US-ASCII) be done for exactly this reason.

Some user agents (e.g. mutt) also allow the user to specify which
encodings to try in what order.  Mine is set to try us-ascii first,
then iso-8859-1, then iso-8859-15, then iso-2022-jp and last the
"fits all solution" utf-8.

-- 
Tobias						PGP: http://9ac7e0bc.uguu.de