CIP-0137 | Fix protocol specifications #1085
Conversation
In order to provide a simpler message invalidation mechanism.
The cold verification key is not strictly necessary in the operational certificate. This allows to simplify the Haskell DMQ implementation and reuse the existing KES signature verification of the Ouroboros consensus implementation.
rphair
added
Update
There was a problem hiding this comment.
@jpraynaud the text parts look understandable to me and I trust the specification changes are authoritative. Others with an interest in the evolution of this proposal, please feel free also to review: @coot @ch1bo @bolt12 @Quantumplation @perturbing
Marking Last Check for next meeting https://hackmd.io/@cip-editors/119 but can be merged earlier if other Network reviewer(s) + other editor(s) are also happy with it.
rphair
added
the
State: Last Check
There was a problem hiding this comment.
I only read the changes and am not involved in the Mithril protocol design. It's great to see that new findings are contributed back to evolve the CIP, which I think is a good practice.
Left a comment though on the operationalCertificate which is treated less a black box now, but might demand some further clarification (for the CIP specification to be self contained).
rphair
removed
the
State: Last Check
4 participants
This PR includes updates of the CIP-0137:
blockNumberfieldttlexpiresAtfield (word32)coldVerificationKeyfield (word32)expiresAtfieldMsgServerDonemessage has been removed from then2c local message notificationmini-protocol(latest version of updated document)