Closed TimoGlastra closed 1 year ago
Discussed on 2023.01.09 AnonCreds Spec Working Group Meeting -- likely agreement, but want to look at the dropping of the ISSUANCEBY* requirements -- needed in AnonCreds or just in the Indy AnonCreds methods.
The issuance_by_*
is not needed technically. It is current used by 2 interfacing functions:
There are 2 ways to update revocation state
issuance_by_*
)Downstream changes to this just means that prover must either provide the old revocation list and old_witness or that they know what variant issuance_by_*
is.
I think this is the summary:
issuance_by_*
is needed by the AnonCreds Method, but not by AnonCreds, so it is OK to leave it out of AnonCreds data structures. By the time the data is passed into AnonCreds for presentation creation, the AnonCreds method must have made sure that the full state of the RevReg has been prepared to be passed in.
An AnonCreds Method that stores deltas (such as Indy) MUST capture from the issuer issuance_by_*
and MUST provide a way for the holder to get that information.
An AnonCreds Method that stores full state (such as Cheqd) need not worry about issuance_by_*
.
I think we should talk about this to the spec., in the data flow setup section -- perhaps with a one-liner and link to this section in the presentation generation section?
Agreed to merge -- 2023.01.23 Meeting.
Updates the revocation registry model to use camelCase and follow the anoncreds implementation.
I've not updated the revocation registry entry related content as that'll be replaced with the revocation list model (in a follow up PR).
I've also updated the wording of the tag to follow that of the credential definition tag, but we may want to rewrite this after an decision has been made regarding #123
Model in AnonCreds RS (same as Indy Shared RS): https://github.com/hyperledger/anoncreds-rs/blob/main/anoncreds/src/data_types/anoncreds/rev_reg_def.rs#L72-L100