Date: Sat, 06 May 1995 14:46:06 -0700 (PDT)
From: Ned Freed <NED(_at_)SIGURD(_dot_)INNOSOFT(_dot_)COM>
The entire reason for doing the security multiparts work in the
first place was to define a security service that works well with
MIME. You now say your solution is intentionally designed so that
it does not meet this primary goal. So why are we even having
this conversation?
As we've covered already, I was under the mistaken impression that 822
compliance implied MIME compliance, i. e. that MIME was a superset. I
thought that during the last time around on this issue (in cypherpunks
anyway) that NSB had affirmed my understanding. While this probably
means that I misinterpreted what he said -- perhaps he merely stated
that MIME was 822 compliant -- this was my honest understanding and it
influenced my decision to readdress this issue.
In any case, during this conversation, you pointed out my
misunderstanding. This caused me to realize that I was not merely
trying to avoid unnecessary encapsulation within MIME, but rather,
avoiding MIME altogether.
BTW, do any of the current proposals deal with 822-compliant messages
which are not MIME compliant? Is 822 dead? Are there plans to strip
out the non-MIME-compliant parts? If someone could point me to an
existing document that addresses this issue, I would appreciate it.
Rick