ietf-smime
[Top] [All Lists]

Re: application/signed vs. application/mime vs. signedData

1997-08-01 14:21:59
At 09:08 AM 8/1/97 -0700, Ned Freed wrote:
Well, I for one disagree with this assessment. The problem with having two
formats is that it then becomes a matter for the sender to choose which
one to
use. And that's simply wrong -- this choice cannot in general be made

        At base, I strongly agree with you.  The reason for suggesting a 
construct
like application/mime was to make sure that we COMPLETELY SEPARATE the
structuring/labeling activites for real data service, like s/mime, from
peculiar transport situations, such as needing to bullet-proof.  That's why
content-transfer-encoding is separate from content-type.

        I am not as sure as Ned about the ultimate and complete failure of 
any/all
bullet-proofing.  My guess is that it can be useful on a small scale and
disastrous on a large scale.  But I am very certain that the the issue must
be dealt with separately from the semantic efforts, as for s/mime.  

        Let's be clear:  There is nothing about s/mime which requires two ways 
of
doing the same thing.

        There ARE differences in particular communications scenarios which need
additional bullet-proofing.  Such bullet-proofing needs to be achieved --
assuming it can work at all -- by one or more separate mechanisms.

        app/mime is an experiment.  there is no question that some gateways will
know about it and mess with it.  At a minimum, the purpose of the
bullet-proofing is to serve as a very clear label to keep one's processing
mits off this thing unless the software knows what it is doing.  

        mulipart has a default handling defined.  it is fine for many situations
and is proving to be a problem for others.  that's what app/mime tries to
respond to.

        if the approach to bullet-proofing used by app/mime works, there needs 
to
be onlyone way of doing it, for a variety of data types.  If it doesn't
work, we need to be able to back out from using it easily.  If we have this
approach to bullet-proofing implemented deeply into each of several
different semantics efforts, we won't be able to back out.

d/
--------------------
Dave Crocker
Internet Mail Consortium                               +1 408 246 8253
675 Spruce Dr.                                    fax: +1 408 249 6205
Sunnyvale, CA 94086 USA              info(_at_)imc(_dot_)org , 
http://www.imc.org

Member, interim Policy Oversight Committee     http://www.gtld-mou.org