Besides, even without this thought, I don't see how your latest proposal
gives an ECC public key holder a mechanism to block transmission with 3DES
(because 3DES sneaks in as an alg. for ECC keys, it may be chosen by the
sender).
my point was that if we have a key preferences flag on the
key, then this is an instruction to the sender's software, i.e.
"strict suiteb ciphers only" - in other words, don't use 3DES.