Re: States and the 8 bit dilemma

• Messages sorted by: [ date ][ thread ][ subject ][ author ]
• Previous message: A. Padgett Peterson P.E. Information Security: "States and the 8 bit dilemma"

On Feb 29,  2:09pm, A. Padgett Peterson P.E. Information Security wrote:
} Subject: States and the 8 bit dilemma
}
} Derek RIP (rites in part)
} >	1) I am sending a complete multimedia message and I want to
} >sign the whole shabang.  In this case the best way to do it is to
} >format the message and sign the whole MIME content (c.f.
} >multipart/signed).  This is message security.

Actually, I don't think that what Derek just described is message
security.  It's object security on an object that happens to be a
message.

Message security, as I understand Michael's original use of the term,
has to do with including some portion of the visible message structure
(for example, the MIME headers of a leaf part) in the same security
"unit" as the object being transmitted.

Take Derek's example and extend the signature to include the header
`Content-Type: message/rfc822', so that along the way someone couldn't
change it to `Content-Type: message/partial' or some such foolishness;
then you have message security.  However, I'm still struggling to find
an example where any undetected change to the headers would be damaging
rather than simply annoying.

If a labeling scheme of the sort MSP provides ("top secret", etc.) were
included in the MIME headers, *THEN* it would be important to prevent
that label from being changed.  That may be the level at which we really
need message security.

} What we really want to do is to create a mechanism whereby such
} authentication is added by default and on-the-fly which also allows
} extraction/verification/ execution in one movement that only requires user
} intervention if something is wrong/missing.
} 
} Biggest change required is from "sign whatever I want" to "sign everything
} unless I say no".

I don't think that's really what we're after.  Consider Nathaniel's
argument that signing/encrypting everything by default creates a
precedent that anything that *appears* to be signed by you *was* in
fact signed by you.  If your key is ever compromised, forged documents
dated in the past can be produced and you'll be hard-pressed to prove
that you didn't originate that such a document.

-- 
Bart Schaefer                     Vice President, Technology, Z-Code Software
schaefer@z-code.com                  Division of NCD Software Corporation
http://www.well.com/www/barts

• Previous message: A. Padgett Peterson P.E. Information Security: "States and the 8 bit dilemma"