Martin:
In particular, I think that CMS encrypted types might be useful here to give
them more flexibility than what they have now, and so they don't eventually
have to reinvent a bunch of CMS. It also might be overkill. Thoughts?
It was an explicit goal to NOT have flexibility in addition to what is
provided by content-encoding negotiation. That is, if
chacha20-poly1305 was considered categorically better, a new
content-encoding would be needed and support for it negotiated using
the mechanisms that HTTP provides.
I do not understand the desire to avoid crypto agility.
Russ
_______________________________________________
smime mailing list
smime(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/smime