Re: credential "recovery" or "reissuance" terminology

My 0.02:

Reissuance can happen for many reasons, not all of which are recovery
scenarios. For example, reissuance might be needed due to a a name change
associated with a life event, because of a typo in the original credential,
or because the issuer needs to replace a compromised issuance key.

I think the terminology for such flows should be identical to that of
original issuance *except* that it has a new step early in the process--
"revocation" of the old credential. So "reissuance" = "revoke old" + "issue
new", whereas ordinary issuance = "issue new".


On Tue, Jul 2, 2019 at 5:16 PM Kim Hamilton <kimdhamilton@gmail.com> wrote:

> This should be straightforward, but I couldn't self-help from the VC data
> model.
>
> Assume subject == holder for now (just in case being different would
> complicate things)
>
> In my scenario, Alice issues a Verifiable Credential to Bob, the subject
> of which is did_b (controlled by Bob). Then suppose for whatever reason,
> Bob loses ability to lose control of did_b*.
>
> In this state, the credential is still valid (hasn't been revoked, hasn't
> expired), but Bob can't prove control of did_b. Do we have special
> terminology for this state and/or verbs involved? E.g. Bob will need to
> "recover" or "request a re-issuance" of a credential from the issuer.
>
> The VC lifecycle doesn't seem to call out separate states/verbs for this,
> which I can understand. Ultimately the flow is the same, and it results in
> an "issuance". But when describing user stories around systems using VCs,
> it's often necessary to call this out, and it would be helpful to use
> standard terminology if available.
>
> * i.e. suppose he's using BTCR, failed to backup the info he needs to
> update did_b, and for simplicity, loses his phone, never made a backup of
> his mnemonic seed. Bad Bob.
>
> p.s. I'm sure I made the question more complicated more than it needed to
> be
>
>