On Mon, 11 Oct 2004 08:35:17 -0700, Jim Fenton wrote:
But the count of MIME-Version headers misses the point. We
have been imprecise in saying that "the need to convert
messages to MIME" is the problem, since a text/plain message
with the MIME information in the headers is indeed a MIME
message, and converting to that wouldn't be a problem. The
question we should be asking is whether it's necessary to
convert non-MIME messages or text/plain messages into
multipart MIME messages.
Or if that definition is imprecise too, please help me with
that.
Jim,
Yes, we need to get this wording precise and correct. And as you
note, it's a tad trickier than might have been obvious (certainly
to me.)
Let me try a line of attack that is more macro-view and less
technical:
We want intermediate MUAs, receiving MUAs and their associated
recipients to be able to continue processing messages without
having to make any changes. In this case, the requirement is far
more strict than it originally was for the introduction of MIME:
"No changes" includes no visible effect on the body of the
message.
d/
--
Dave Crocker
Brandenburg InternetWorking
+1.408.246.8253
dcrocker a t ...
www.brandenburg.com