- From: Dave Longley <dlongley@digitalbazaar.com>
- Date: Mon, 26 Jun 2017 14:09:36 -0400
- To: Steven Rowat <steven_rowat@sunshine.net>, public-credentials@w3.org
On 06/26/2017 01:34 PM, Steven Rowat wrote: > On 2017-06-26 9:27 AM, Manu Sporny wrote: >> * The number of options for ROLE_C has become so large that it will >> most likely lead to bad polling results. I suggest that we start >> aggressively culling the ROLE_C list before the vote starts >> tomorrow. We should get some strong arguments against roles that >> you feel should not be in the running. > > Here's my attempt to cull new Role C (the Holder/Presenter/... list) > : > > -1 PROVER This seems in the wrong role. IMO it's closer to what role > D does by checking the claim. Or even Role B, who checked various > facts before issuing the claim. I can't see it fitting in what > Holder/Presenter does. I'm not an advocate for this name either, but my understanding of the reasoning for it is: this Role is responsible for providing proof to the Inspector that the claim was made. That proof may be somehow found in the claim itself or it may be performed via some zero knowledge challenge. I'd prefer to remove it as well, because it seems specific to certain proof protocols. I don't think it really captures the essence of Role C and I believe other terms could do better whilst encapsulating the "Prover" aspect of those proof protocols. Regarding CLAIMANT: I'm a -1 on Claimant for reasons stated before -- I think it introduces confusion because multiple Roles are effectively making claims in the system. Role C does make a claim, which is that "The Issuer said X about the Subject." And that claim *is* what is verifiable. However, when we talk about the data model and the claim we're actually modeling, we are talking about X. We're talking about how we model what the *Issuer* has claimed about the Subject. The claim made by Role C is merely implied by the act of sharing. So I think it is confusing to use the term Claimant at all here -- as we've got a concrete claim that we're modeling (made by the Issuer) and an implied, more ephemeral claim that results from the sharing of that Issuer's claim with an Inspector. As noted before, some groups have called the Issuer's claim an "attestation" or "assertion", but as this is just a synonym for "claim", I don't think it actually promotes understanding. -- Dave Longley CTO Digital Bazaar, Inc. http://digitalbazaar.com
Received on Monday, 26 June 2017 18:10:04 UTC