Hi,

I have a few questions/remarks about XEP-0384:

- The examples mention signed pre-keys. This isn't part of the Olm protocol
AFAICT, I only saw it in the original double ratchet spec. Is this a
remnant of an older version of the XEP? Should this be dropped, or is the
plan to use X3DH eventually?

- The XEP uses AES128-GCM for encryption, which seems different from Olm
(which uses AES256-CBC). I know very little about crypto, but is there a
reason for going with AES128 instead of AES256? I'm asking, because on
first sight, it seemed easier to find support for AES256-GCM (e.g. in
libsodium/nacl) than for AES128-GCM.

- I can't find a reference in the XEP on how the ratchet algorithm works in
multi-user settings (there's only a reference to Olm, which AFAICT only
works for single users). Is this where megaolm steps in? Should there be a
reference to a protocol?

- The Device ID being a randomly generated 32-bit integer seems out of the
ordinary? Not that the chances of collisions in a big MUC are *that* big,
but would a GUID be more consistent with some previous XEPs? Should
something be said about what to do when recipient device IDs collide?

- Cosmetic: camel case is used for tags/attributes, whereas I believe
kebab-case is more consistent with other XEPs?

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

Reply via email to