On 31 March 2017 at 15:40, <[email protected]> wrote:

> It just skips the conversion part altogether, which seems to be the
> central point of discussion and is not widely implemented in any direction


Interesting! I believe there *is* the open question whether using the same
key for signing and encrypting is a good idea, which this solution would
avoid entirely.

Libsignal users could reach into the library and use convert_mont to
publish the same key for signing as for encrypting, avoiding the need for
them to regenerate identity keys. This just leaves the receiving path of
libsignal to be patched to take an arbitrary Ed25512 key instead of
converting it from the X25519 key first (i don't even think the validation
needs to be changed to use EdDSA, since Ed25512 signatures are also valid
XEd25512 signatures according to the spec?).

The fingerprints would change, but those are typically only used in a
short-lived scenarios anyway?

Remko
_______________________________________________
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: [email protected]
_______________________________________________

Reply via email to