Discussing the opportunity of allowing shared_secret extraction while computing the recipients' output keys for some transaction.
To be more precise we could either expose the ecdh_shared_secret that is computed from the recipient B_scan and the partial_secret, which is a PublicKey, or a map of all the actual secret that are computed with calculate_t_n. First one let the downstream protocol responsible for implementing its own shared_secret, opening the possibility for deriving different keys from the same ecdh_shared_secret, while the second one could maybe be easily implemented by changing the return value of generate_recipient_pubkeys.
That would be useful for some niche and slightly deviant use cases I'm working on, where silent payments transaction are used as a notification + key exchange scheme for higher level protocols. Since that's arguably niche and has unknown security implications I'd rather discuss first if it should make it into the lib at all, or be left up to the bastard messing with this stuff.
Discussing the opportunity of allowing shared_secret extraction while computing the recipients' output keys for some transaction.
To be more precise we could either expose the
ecdh_shared_secretthat is computed from the recipient B_scan and the partial_secret, which is aPublicKey, or a map of all the actual secret that are computed withcalculate_t_n. First one let the downstream protocol responsible for implementing its own shared_secret, opening the possibility for deriving different keys from the sameecdh_shared_secret, while the second one could maybe be easily implemented by changing the return value ofgenerate_recipient_pubkeys.That would be useful for some niche and slightly deviant use cases I'm working on, where silent payments transaction are used as a notification + key exchange scheme for higher level protocols. Since that's arguably niche and has unknown security implications I'd rather discuss first if it should make it into the lib at all, or be left up to the bastard messing with this stuff.