Re: Verifiable Claims Telecon Minutes for 2016-02-09

On 02/16/2016 09:23 AM, Daniel Burnett wrote:
> 
> 
> On Mon, Feb 15, 2016 at 4:10 PM, Steven Rowat <steven_rowat@sunshine.net
> <mailto:steven_rowat@sunshine.net>> wrote:
> 
> 
>      > B declares a credential. B is a Declarer.
> 
> 
> 
>     To add to my own post, another option here is:
> 
>     B declares a credential about A. So:
> 
>     B is a Declarer.
>     A is the Subject.
> 
>         A takes the credential to C. C is an Acceptor of the credential.
> 
> 
>     Thus Declarer, Subject, Acceptor.
> 
>     An advantage of this would be not having '-or' or '-er' endings on
>     all three, which might make it easier to parse, understand, and
>     remember the documents and process.
> 
>     The Subject, which is what it's all about, is the one that stands
>     out as different, and is central to the process.
> 
> 
> +1 for Subject.  Not only does it work well when humans are the Subject,
> it works also when non-humans are.  For example, in a claim that no
> antibiotics were found in the cow just before it was killed for meat,
> the cow is the Subject of the claim.  Or in a claim that a car was in
> exactly two accidents in the state of California, the car is the
> Subject.  It has the advantage of not implying whether the Subject is
> the initiator of the claim or whether the claim was made about the
> Subject independent of any action on the Subject's part.

I'm also +1 for Subject (at DB we've actually debated suggesting that
term for some time now). The "subject of the claim" is very clear to me,
but the push back on "subject" has been that it seems like it requires a
more technical view than "holder" does. I don't know if that's true.
Another thing to point out is that we may actually want the "holder"
terminology *and* the "subject" terminology.

In the CG work, we've modeled an "identity credential" as having its own
properties, only one of which is the actual "claim" being made. The
other properties include things like issuer, issue date, and so on. The
"subject of the claim" (the object pointed to by the claim property in
the model) is fitting here. However, we also want to be able to talk
about the "holder" of that credential. That may not be the same entity
as the subject; certainly it is not in the case of an object with no
agency such as an inanimate food product.


-- 
Dave Longley
CTO
Digital Bazaar, Inc.

Received on Tuesday, 16 February 2016 15:46:50 UTC