On 2016-06-11 01:17, "Jim Schaad" <[email protected]> wrote:
>> >>Section 5.2 >>"The COSE_Encrypt1 encrypted structure does not have the ability to >>specify >>recipients of the message. The structure assumes that the recipient of >>the object >>will already know the identity of the key to be used in order to decrypt >>the >>message. If a key needs to be identified to the recipient, the enveloped >>structure >>ought to be used.” >>One type of compact encryption formats we have looked at has been, >>essentially: [key identifier, cipher text] (disregarding IV etc.). We >>would like to >>use COSE_Encrypt1 which essentially is: [algorithm, cipher text] >>(disregarding IV >>etc.). Making algorithm optional is covered in Appendix A. But I read >>the text >>above as using this format with a kid header is not possible, is that >>the intention? >>I can maybe understand this from a level/layer structure point of, but >>not really >>from a security point of view. > >The choice of the word 'ought' was very deliberate. The intention is to >say that this is what it is believed should be done, but it is not by any >means a statement of requirement. You can violate this without being in >violation of the rules, but you should really think about why you are >doing it. Some more thoughts on this. I understand this is not normative. It sounds like you say it is not recommended (SHOULD NOT) but is allowed (MAY) given certain important security considerations are made. - Why not formulate it like that? - I didn’t find any security considerations specifically on this, but may have missed them. If they are not already in, I think some considerations should be included. Comparing with DTLS, there is essentially a Sender ID (disregarding IV etc.), no algorithm and no enveloped structure. I assume that DTLS fulfils those security considerations, so maybe that can serve as a good example? Since this may be a special case of no algorithm, some text about the case in question could be included in Appendix A. Göran _______________________________________________ COSE mailing list [email protected] https://www.ietf.org/mailman/listinfo/cose
