Re: [COSE] MAC with no recipient structures

["Jim Schaad" <[e-mail]>]

I just read through the section referenced and did not see any reason for C and CAM to have access to the content.  I will try and read the entire document later today as that might describe the reasoning.

However, you are saying that it is ok for incorrect content to be in this message as far as C and CAM are concerned?

Jim


> -----Original Message-----
> From: Olaf Bergmann [mailto:[e-mail]]
> Sent: Saturday, November 14, 2015 1:26 AM
> To: Jim Schaad <[e-mail]>
> Cc: [e-mail]
> Subject: Re: [COSE] MAC with no recipient structures
> 
> "Jim Schaad" <[e-mail]> writes:
> 
> > People keep telling me that they want to have a version of MACs that
> > do not have a set of recipient information attached so that they can
> > do direct MACs.  I keep asking for a use case where this makes sense.
> > In all of the use cases that I have been presented so far, a better
> > answer is going to be to do an AEAD encrypted item rather than a MACed
> item.
> 
> An example where this is useful is given in Section 3.7 of draft-bergmann-ace-
> dcaf-cose [1]. Here, the recipient structure is empty because CAM and C must be
> able to read the payload. Technically, you can make this work with an encrypted
> AEAD item as well but then, the payload information entirely goes into the AAD
> part because CAM and C must not know the secret that is being used for
> generating the MAC tag (they just forward the data to S).
> 
> Of course, this is a requirement only in the cross-domain authorization scenario
> where C and S can be constrained and can have distinct owners.
> Where this is not considered an issue, you can easily go with COSE_encryptData
> and AEAD.
> 
> [1] https://tools.ietf.org/html/draft-bergmann-ace-dcaf-cose-00#section-3.7
> 
> Grüße
> Olaf