W3C home > Mailing lists > Public > public-credentials@w3.org > August 2021

[MINUTES] W3C Verifiable Credentials VC-EDU Call - 2021-08-30 11am ET

From: Kerri Lemoie <klemoie@concentricsky.com>
Date: Tue, 31 Aug 2021 13:33:59 -0400
Message-Id: <A96912E2-1EC7-4724-86DC-4109C3FEC70E@concentricsky.com>
To: W3C Credentials CG <public-credentials@w3.org>
Thanks to Marty Reed for scribing this week! The minutes
for this week's  telecon are now available:

https://w3c-ccg.github.io/meetings/2021-08-30-vc-education/

Full text of the discussion follows for W3C archival purposes.
Audio from the meeting is available as well (link provided below).

----------------------------------------------------------------
Minutes for 2021-08-30

Agenda
https://lists.w3.org/Archives/Public/public-credentials/2021Aug/0380.html
Topics
		• Agenda Review
		• IP Note
		• Call Notes
		• Scribe Selection
		• Introductions & Reintroductions
		• Announcements & Reminders
		• Anthony Camilleri - Europass as VCs
Organizer
Kerri Lemoie, Anthony Camilleri
Scribe
Marty Reed
Present
Kerri Lemoie, Anthony Ronning, Phil Long, Steve Traugott, Marty Reed, Wes Teter, Steve Gance, Anthony Camilleri, Thierry Thevenet, Mark Miller, Tzviya Siegman, Deb Everhart, Matt Lisle, Sheryl, Sharon Leu, Juan Caballero, Jim Goodell, Simone Ravaioli (Digitary), Dmitri Zagidulin, Stuart Freeman, Nate Otto, David Ward, Adam Lemmon, Simone Ravaoli, Taylor (LEF), Kayode Ezike, David Chadwick
Audio Log
audio.ogg

Topic: Agenda Review
Topic: IP Note
Kerri Lemoie: https://www.w3.org/accounts/request
Kerri Lemoie: https://www.w3.org/community/about/agreements/cla/
Topic: Call Notes
Kerri Lemoie: https://w3c-ccg.github.io/meetings/
Kerri Lemoie: http://irc.w3.org/?channels=ccg
Topic: Scribe Selection
<phil_l_(p1)> I'll scribe
<phil_l_(p1)> Ok - thanks Marty!
Marty Reed is scribing.
KL: introductions reintroductions
Topic: Introductions & Reintroductions
DE: intro of Credential Engine
KL: Anthony is going to be introducing EuroPass
Topic: Announcements & Reminders
KL: Every Tuesday weekly CCG meeting
<kerri_lemoie> Work item to write a VC Guide for Developers:
KL: October IIW 12th-14th
Kerri Lemoie: https://www.eventbrite.com/e/internet-identity-workshop-iiwxxxiii-33-2021b-tickets-160257990965
Topic: Anthony Camilleri - Europass as VCs
KL: intro Anthony
AC: introducing digital credentials part of EuroPasss
AC: major policy project of the commission, approximately 2MM accounts and 10k new per day
AC: european standard comes out of leg. 2018 mandating a Euro system for digitally signed qualifications
AC: starting with design prinicipals and standards to address all levels of education Prek-University
AC: align with Euro Recognition Instruments, harmonizing education systems in Europe
AC: Capture Formal and Informal, learning is learning
AC: applicable to the whole course lifecycle
AC: interoperable, free, and open source
AC: legal value across the EU
AC: no standard fully reflected lifelong learning, therefore develop meta-model ELDM
VC model was a good foundation to start with
AC: something you have gained related to learning activitives
AC: achievement is something you've earned, learning outcomes, knowledge, autonomy and responsibility
AC: including entitlements
AC: Lifecycle Concepts: specifications, learning opportunities, credentials describing a claim
AC: specifications are interrelated, building on one another, occupation, course, lesson plan
AC: Learning opportunity specification, add individual information to create a credential
AC: Credential metadata- Issuer data and credential type, Claim - Person ID (10-15 different identifiers) set of legal identifiers defined by EU for signature, and action
AC: VC XML currently moving toward an RDF/JSONLD model
P1: question on entitlement
AC: entitlement = right to study, membership of an org as an example
AC: start from Open Badges V3 spec
<juan_caballero_(spruce)> (sorry i was confused, was switching devices :D )
<kerri_lemoie> All good @Juan :)
AC: the majority of our credentials are highly complex 60-70 claims per credential
AC:things we can use - type, achievementType
AC: TWCU - badgeclass and resultsDescription
<nate_otto> When you say a credential has 60 claims in it, area these 60 described achievements, each of which might have entitlements etc? Packaged together like a transcript? (For next pause)
AC: TWCU - no nesting relationship between badgeclass and resultdescription
AC: TWCU - everything has a URI
AC: the idea of creator inside credential subject is useful to reference entities outside the credential
AC: Issues to Resolve - cannot support multiple claims
AC: proposal to solve - keep structure and create an array of claims
AC: includes multiple types of claims within the array
AC: ITR - credential has a type at the top level
AC: Proposal - ass a credential type as a property
Add not ass
<juan_caballero_(spruce)> credential title as arbitrary string/local alias or specified?
AC: within VC there are no properties to describe the credential as a whole
AC: intrinsic part has to be the visual representation
AC: proposal - add displayproperties tag
<juan_caballero_(spruce)> i'm assuming that `"id":"https://website-that-displays-credential.example.com"` really means `"id":"https://website-that-displays-credential.example.com/credential/{UUID}"`, right?
AC: put the HTML directly in line within the credential
JC: question around credential title
AC: the credentialtitle is a simple text string
JC: like an alias string
JC: UUID in URL?
AC: ID of the credential is a UUID, link to a visual representation can be a URL in the description
P1: link pointing to an IPFS location or the like?
<juan_caballero_(spruce)> CIDs not UUIDs :D
AC: in our implementation the entire HTML will be in the credential
<dmitri_zagidulin> (one could also use hashlinks, to that HTML)
<juan_caballero_(spruce)> ^
AC: adding the ability to add an outside reference link
P1: alternative hashed link?
<anthony> (this one I'm taking back with me :-))
KL: your example shows a claim with an array of claims, how do you handle the multiple achievements in the claim?
AC: description of diploma supplement
<kerri_lemoie> Anthony's example: https://hackmd.io/D2b4lJTXTgOG1ss4B3payA
AC: want to be able to represent single expresssions as well in the same model
NO: separability of claims?
<deb_everhart> Nate asked my question
<phil_l_(p1)> Might the signature method be able to 'separate out a credential?
AC: massively complex question, technically separating is on the roadmap, other part is to have individual claims and stackability of those claims
P1: self-assertion of a credential?
AC: self-assertion is not within the model
<juan_caballero_(spruce)> "issuer": "did:example:some_random_guy"
AC: use HROpen as a self-assertion model
<deb_everhart> is that the resume portion of HR Open?
KL: how will evidence be used?
AC: evidence is quite controversial
AC: current plan is for evidence to follow OpenBadges
https://www.w3.org/TR/vc-data-model/#example-23-usage-of-the-evidence-property
AC: actively talking about semi-universal wallet
<juan_caballero_(spruce)> universal-ish
AC: OpenBadges and EuroPass creds in the same wallet
<dmitri_zagidulin> @Juan: +1 to 'universal-ish', great technical term!
<phil_l_(p1)> Does that mean that there is in the current Europass wallet the only type of credential that can be held?
JC: evidence is inside the credentialsubject?
KL: evidence in an OB is different than VC
<nate_otto> OB did intentionally model evidence as a claim about the recipient in our v3 proposal.
<phil_l_(p1)> a video file
KL: can be part of the achievement
<simone> "Uni(contro)versial Wallet" ? (apologies, just trying to match Juan's sharp commentary ;-)
AC: surprised no one has questions about multiple claims?
AC: is this a valid way forward?
DZ: recommend VCs should be short, but realizing there are legislative requrements
<juan_caballero_(spruce)> or resignation
<taylor_(lef)> @JC, Kerri is the antidote to willful OB ignorance. I'm guilty as well :)
DZ: you're hearing acceptance
KL: we're seeing alignment between OB and CLR
<dmitri_zagidulin> I think everyone is hoping that the two approaches will be interchangeable
KL: many cases the credential should be simple, but many that require more complexity
DE: is it as simple as a both/and?
<dmitri_zagidulin> meaning, if you have a bunch of short credentials, that you can bind them into a single larger one. Or, if you have a large multi-claim credential, you can split out claims (using BBS+ etc)
DE: can it be simple and then referenced from the more complex?
<dmitri_zagidulin> (or, if not able to use BBS+ or similar tech, it could be as simple as - ask your issuer for BOTH a large "transcript", and a specific skill credential.
DE: OBv3 and CLR will both support multiple claims?
Phil Long: +1 Dimitri
<anthony> @dmitri that last one is what happens in pratice in many institutions
KL: OBv3 is still a single achievement
<dmitri_zagidulin> @Anthony - makes sense!
AC: issue multiple small credentials and then combining them?
DE: yes oversimplified
<deb_everhart> one important value of that approach is to give people their achievements sooner, along the way to the larger credential
AC: in the transcript you could do a number of small credentials, but there are many use cases that still require multiple claims in a single credential
Kerri Lemoie: +1 Deb
P1: OBv3 as a building block is a great value, complex credentials have a layering with relationships between those claims and sequence of those claims matter
P1: metadata of the relationships is required in the more complex credentials
P1: complex credentials work ahead is the metadata patterning
AC: agreed
<kerri_lemoie> :)
AC: relationships between claims is a key issue, we must describe both approaches
AC: issuer issuing a credential with other relationships and how do you do that?
P1: some organizations wish to have a set of claims as a single package and allow those claims to be expressed as a new set of credentials
AC: topics for next week, multiple languages, guidance on DIDs vs PII, Labeling Skills/learning Outcomes, Micro-credentials and stackability in a VC
<juan_caballero_(spruce)> thanks anthony and marty!
KL: close 





Received on Tuesday, 31 August 2021 17:34:14 UTC

This archive was generated by hypermail 2.4.0 : Tuesday, 31 August 2021 17:34:17 UTC