<burn> scribenick: DavidC
RezaS can you join the voice call?
First week of April
Location is San Francisco
Details will be sent out by email
<RezaS> David yes I am joining right now
Manu will work with Chairs to see if DB needs to bring projector for IIW
<RezaS> Just joined
RezaS introduced himself. Is a PG at York Uni, Canada
His research topic is identity management
<burn> Data model issues: https://github.com/w3c/vc-data-model/issues
<agropper> Having trouble dialing in.
there are 5 new issues that need to be assigned
<dlongley> RezaS: You may also be interested in joining the #ccg channel and the call that occurs right after this one in that group.
Issue 123
<manu> DavidC: Takling about issue --- https://github.com/w3c/vc-data-model/issues/123
<dlongley> DavidC: It's the fact that the type is not specified until you look into the type parameter. When the type is embedded in something it is specified.
<dlongley> DavidC: I give an example, a credential -- a credential is not set to be a credential then it is set to be a credential.
<dlongley> DavidC: What I'm proposing is that the types be specified whether they are embedded or not.
<gkellogg> I’ll take it.
<dlongley> i thought we always specified the types already :)
Issue 124
Privacy retention times
Issue 125
Anonymous credentials
Manu assigned to 124 and 125
Issue 126
dlongley assigned
Liam asked to do the assignment for us
<agropper> that was me Adrian Gropper
Callin user 5 was booted out because he/she did not identify themself
Issue 127 taken by gkellog
<agropper> apologies - the call assignment protocol here is different than the CG
<dlongley> DavidC: Dave Longley and I reached an agreement that we wanted to add more examples and text to make it clear to the reader that things can be lists or must be, etc.
<dlongley> DavidC: It just requires a PR to put some appropriate text into the data model.
<dlongley> DavidC: I'll start the PR.
<Zakim> manu, you wanted to discuss status update (or lack thereof -- but test suite plan!)
<liam_> [ to send pull requests on github to w3c specs you may need to go to https://www.w3.org/Help/Account/ and "edit your account information" and then on the left scroll down to Connected accounts, and connect your github account ]
Manu was at rebooting last week so did not have time to work on his issues
Nevertheless progress was made on delegation last week
Suggestion is to use object capabilities
<dlongley> to summarize the idea: Verifiable Credentials "entitle" you to get an (object) Capability -- which can then be delegated. If the entity that provided the capability does not want delegation to happen, they can restrict the capability (in arbitrary ways) before handing it over.
Manu said that a number of companies require VCs but are not in our group yet
Manu. We are 50% through our time so need to get our model out for review if it is to go through all the standardisation phases in time
<gkellogg> I’m not likely to implement for some time, so don’t count on me.
Manu. We need implementations now
<dlongley> DavidC: My question is to Manu, I wasn't there last week. But the fine dividing line between object capabilities and verifiable credentials... they are the same thing to me, I'm wondering why they need to be differentiated.
<dlongley> ChristopherA: My comment was that one of the big barriers that I'm running into ... and it may be contributing to the perception that VC is not complete is barely begun, etc. In absence of signatures, VC are not very valuable. When you add in signatures it begins to reveal things that change the semantics.
<dlongley> ChristopherA: I've had some discussions with Manu and others to formalize that. I know we can't, it's not within our charter, but it's becoming an obstacle. Maybe some people participating in this and not in the signature part -- we may need some help with the signatures.
<dlongley> cwebber: I got on the queue to respond to David Chadwick. About VC vs. Object Capabilities. They are not the same thing in this sense. VC do say -- they are badge like, saying things like "this person is a SysOp, etc." If we end up focusing on "this person is this", that's not a capability that's an Access Control List.
<dlongley> cwebber: I think we saw we need more data modeling to create an Access Control List. And many are confused about ambient authority and confused deputies. Both of these technologies are interrelated and need each other. When you have a human that says whether you are qualified to lease an apartment or get a job, etc. they check for job references, uni degrees, etc. Then that results in you getting an apartment key to enter the building.
<manu> cwebber: You may have verifiable credentials that break into other people's machines... prove that's true, revoke that privilege, high security company, not acceptable for our particular work case.
<manu> DavidC: I think you're talking about the same concept
<dlongley> cwebber: What we've been talking about is that the human reasoning type things end up happening -- those depend on VCs.
<cwebber> a role is equivalent to an ACL group
<manu> DavidC: Generic or indirect action or not... you said it's like an access control list, but when you go to RBAC, don't see difference between role and a capability. You present it and you have a role in the system, got some in built knowledge. Capabilities say you can open the door, doens't know what "open the door" means.
<manu> cwebber: You have to understand what "open the door means", but you hold the key to open the door... sounds like a subtle distinction, but it's important.
<manu> burn: This is one of the things that needs to change in this group - we can't have this sort of discussion on the call -- please create an issue and we'll discuss it.
<manu> The issue is being tracked here: https://github.com/w3c/vc-data-model/issues/72
<manu> burn: Discussion is important, but not during status check.
<manu> s/sorry//
Issue 118
JoeAndrieu is assigned to this, but could not assign himself
<dlongley> JoeAndrieu: please see my comments on that issue as well.
Issue 117
This has been completed and will be closed
Issue 112
<dlongley> +1 to close it
This can be closed with no action needed
<dlongley> i weighed in with a +1 :)
<dlongley> https://github.com/w3c/vc-data-model/issues/112
Issue 111
https://github.com/w3c/vc-data-model/issues/111
Noone assigned to this yet
<dlongley> DavidC: This was an action in the lifecycle of VC but not in the data model. Do we need this to be added?
<manu> DavidC: Joe said he thought it was out of scope... whether it should be in or not.
<manu> burn: Are you willing to monitor and give us an update on its status.
<manu> DavidC: Yes
<cwebber> sorry, forgot to re-mute
Assigned to DavidC
Issue 119
https://github.com/w3c/vc-data-model/issues/111
<dlongley> https://github.com/w3c/vc-data-model/issues/119
<burn> We are now on 107
<dlongley> Issue 107
<dlongley> https://github.com/w3c/vc-data-model/issues/107
<manu> https://github.com/w3c/vc-data-model/issues/107
<dlongley> DavidC: No progress since the last meeting.
<dlongley> DavidC: I apologize will get back to it later, no progress.
<dlongley> Issue 106
<dlongley> https://github.com/w3c/vc-data-model/issues/106
<dlongley> DavidC: There is a PR for this one.
<dlongley> DavidC: There have been a couple of comments on the PR and I updated it based on the comments. Nothing since then. I think people have been away at RWoT.
<dlongley> DavidC: There's a PR and we need more review to see if any more changes are required.
JoeAndrieu: nothing to report on use cases
<ChristopherA> Who else is coding signatures?
Is there anything from RWoT to tell us?
JoeAndrieu: will create an issue about the ID used in the profile field
It currently refers to the holder but should be to the profile object
<Zakim> manu, you wanted to suggest issue processing procedure... to underscore what DanB said.
<dlongley> +1 for specific spec text in issues.
<dlongley> DavidC: I think that issue about the profile ID has been raised already. I did have some long discussions with Dave Longley.
<dlongley> JoeAndrieu: That's right, that came out of that conversation, Dave's response to you -- it doesn't fit the overall pattern for ID fields.
<dlongley> DavidC: It may be that an issue isn't needed and we can resolve it in the current issue.
<dlongley> JoeAndrieu: I'll look and see if that's the right way to do it.
<dlongley> burn: Thanks to everyone, this was a good meeting, we made progress. Thanks in particular for work outside of this call.
<burn> rrssagent, draft minutes
This is scribe.perl Revision: 1.152 of Date: 2017/02/06 11:04:15 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00) Succeeded: s/Manu will bring projector for IIW/Manu will work with Chairs to see if DB needs to bring projector for IIW/ Succeeded: s/after this discussion please// Succeeded: s/sorry// FAILED: s/sorry// Succeeded: s/apologies// Succeeded: s/yes we're diving down a rathole :)// Succeeded: s/I'm withdrawing myself from this conversation :)// Present: Benjamin_Young Chris_Webber Christopher_Allen Daniel_Burnett Dave_Longley David_Chadwick Gregg_Kellogg Joe_Andrieu Liam_Quin Manu_Sporny Reza_Soltani Ted_Thibodeau Adrian_Gropper Regrets: Tzviya_Siegman Found ScribeNick: DavidC Inferring Scribes: DavidC Agenda: https://lists.w3.org/Archives/Public/public-vc-wg/2018Mar/0002.html WARNING: No date found! Assuming today. (Hint: Specify the W3C IRC log URL, and the date will be determined from that.) Or specify the date like this: <dbooth> Date: 12 Sep 2002 People with action items: WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option. WARNING: IRC log location not specified! (You can ignore this warning if you do not want the generated minutes to contain a link to the original IRC log.)[End of scribe.perl diagnostic output]