Daniel Bluhm

Declaring some of my assumptions as preamble to a question: - VCDI format supports many credential formats - First target format to support is AnonCreds in W3C format As currently...

@gvelez17 I was hoping you might have recommendations for where the right points in the implementation would be to extend later to support other credential formats and signature types.

> > @gvelez17 I was hoping you might have recommendations for where the right points in the implementation would be to extend later to support other credential formats and signature...

The fix in the diff you linked seems like a reasonable approach to me!

Resolved by #2824; further discussion on #2834 for the general case of setting User-Agent header and whether we want to make the value used configurable.

@genaris just FYI, with these changes, ACA-Py and AFJ can now successfully complete OOB + DIDExchange with did:peer:4 (or AFJ emitting did:peer:1 and ACA-Py emitting did:peer:4, I believe). I commented...

I haven't gotten this quite right yet. Checking DID Exchange backwards compatibility with earlier versions of ACA-Py, I get this error: ``` alice_1 | Traceback (most recent call last): alice_1...

I've spent a fair amount of time sifting through the behavior of ACA-Py as it relates to handling messages from protocol versions different than the explicitly supported versions. What should...

>Within a given major version of a protocol, an agent should: >- Respond to a minimum supported minor version, defaulting to "0". > - An agent SHOULD keep minimum supported...

I might be splitting hairs :sweat_smile: I think the ideal would be to only "implement" a single minor version of any major version of a protocol. Honestly wondering if DID...