The W3C Credentials Community Group

Verifiable Claims and Digital Verification

Go Back

CCG Verifiable Credentials for Education Task Force Telecon

Minutes for 2020-08-10

Agenda
https://lists.w3.org/Archives/Public/public-credentials/2020Aug/0031.html
Topics
  1. Introductions & Reintroductions
  2. Modeling VCs for education
Organizer
Heather Vescent, Wayne Chang, Kim Hamilton Duffy
Scribe
Stuart Freeman
Present
Kim Hamilton Duffy, James Chartrand, J. Philipp Schmidt, Stuart Freeman, Juan Caballero, George Artem, Robbie Jones, Dmitri Zagidulin, Kerri Lemoie, Jeanne Kitchens, Nate Otto, Nathan George, Jim Goodell, Adam Lemmon, Chris Winczewski, Simone Ravaoli, Taylor Kendall, Tzviya Siegman, Phil Archer
Audio Log
Did we start to use jitsi?
Juan Caballero: Thanks stewart!
Stuart Freeman is scribing.
Kim Hamilton Duffy: Introductions and re-intros

Topic: Introductions & Reintroductions

James_Dz: with blockchain group at kaiser permanente
Nate Otto: Present 50% (double booked today not on audio)
George Artem: George Artem from Seatte, WA [scribe assist by Kim Hamilton Duffy]
Nathan George: UW grad student, trying to do a vc pilot
Jeanne Kitchens: With credential engine

Topic: Modeling VCs for education

... updates, context: building a document with the goal of clarifying fitting existing schemas to the vc format
... document pulls together examples of best practices pilots
... we're an incubation org not a standards org, but we work with the standards orgs to ensure we're going the correct direction
... document outlines types of credentials people want to create
... one interesting area is what to do with xml
... example 4 shows transcribing to json, but is currently in xml
... why not as easy as just mapping to json, xml format for conforming to European legal standards
... may take months to year to get legal approval for json, could just do xml to conform to current legal reqs
... we don't want to wait for legislature to allow other signature suites
... extend vc data model to express support for xml mapping including any necessary context needed in w3c
... credentials community group is the home for the vc data model
... not a general recommendation for any group serializing xml
... a case by case decision, to make it easier for ppl with strict xml requirements
... hoping that changes from w3c will allow edci to continue using xml formatted credentials
... as well as other groups such as scholarly publishing
... moving on to open discussion topics
... generalization of vocab, discussion of issuance roles and impact,
... ler wrapper standardization
JimKelly_PESC: launched as ilr recommended practices, work is now underway
Simone Ravaoli: Ilr working group meeting thursday 11am eastern
... beginning base document for the workstream, build on w3c work
... wrapper doc and doc we're discussing today
Nate Otto: @Kimhd I will join the audio stream for this meeting then
... join call on thurs if you are interested
... will go through base doc and governance issues
Kim Hamilton Duffy: Will assume ieee is home for ler going forward, and try to keep our docs up to date
... we are standards integration and getting pilots off the ground
... ieee is more a combination of standardization (taking our work as examples) as well as concerns around governance
... next discuss generalization of vocab
... context, pilots want to work with the standard, but aren't familiar with the existing work
Robbie Jones: I can see the agenda
Nate Otto: Catching up, we're on page 2 of this document: https://docs.google.com/document/d/1pt-VNnjoYgl23Mlu0Tjyax5RgANPBfDijERz0SNYfSo/edit
... want to integrate with existing standards groups like open badges
Nate Otto: There's a common pattern accross these groups
... defined credential or badge can be issued to multiple people over time
... can we use a property across use cases
... "degree" seems limiting
... could use something like "badge" or invent something new
Kim Hamilton Duffy: "Has achieved" seems like it fits well across standards
... could compare ler
Kerri Lemoie: Is this to say "has achieved this type of credential"?
Nate Otto: Not type, you could say has acchieved bachelors, but we want to say has achieved bachelors of (subject) from (institution)
... could also map to skills
Jeanne Kitchens: Ctid does not indicate a cred has been issued, more like a unique product identifier
Nate Otto: I asserted "A credential has been assigned an ID by Credential Engine" as the definition of "CTID"
Kerri Lemoie: +1 To bridge
Kim Hamilton Duffy: Credential engine also has a registry of credential types, we're talking about helping bridge between these registries and actual degrees
JimKelly_PESC: like "has achieved" as a generalization, but where does it live? in the wrapper?
Kim Hamilton Duffy: Vc itself is just the wrapper, ler/ilr is a lot like this
... the terms and type definitions will likely live at the standards body, this is the bridge
Nate Otto: There's a separation of wrapper standard and payload standard
... can use flexible range
... can we come up with a consistent way to represent across standards
... could start by trying pilots with "has achieved"
... prove that this has value in the market then bring it to the standards body that we are bridging in the pilot
... not forcing another standards body to adopt, but provide open standard so they can
Kim Hamilton Duffy: Nate do you want to respond?
Tzviya Siegman: Sounds like you allow other standards to implement this so they can be compatible, that could be a heavy lift
Nate Otto: If we can prove the value of the approach the standards may follow
... opportunity for standards bodies to align with wrapper and payload to prove interoperability
... allow machine understanding of the credential
... learn enough to be able to get to machine understanding
Kim Hamilton Duffy: Nate I'll give you the floor to respond
Phil Archer: Clarification, looks to be proper use of the "extensions" property of badges
Nate Otto: Don't think this quite "extensions"
... can duplicate data in properties to layer in w3c properties
... not enough to meaningfully validate
... probably could use the "extensions" but it's probably not enough to do this
Kim Hamilton Duffy: Defining "has achieved" in open badges, we could pass this along to them to see if they'd adopt
Tzviya Siegman: In badging, issuing organization may not be the same kind of organization as a credential, can we assert accreditation by university?
Nate Otto: Open badges infers that the definer of the achievment is the only one who can issue
... there could be use cases for making more generic achievments that can be issued by someone other than the creator
JimKelly_PESC: this feels like just one of many characteristics that may be required for true interop
... could come up with a whole set that need to be shared across payload specs
Kim Hamilton Duffy: That's true, don't want to reinvent the wheel
... use best practices from different standards bodies, would love to just use an existing set of terms if they exist
... as a pilot group we have to id those and turn them over to standards groups
... like proceeding with "has achieved"
... proposal: these are using json-ld, maintain a set of work-in-progress schemas
... dcc has a playground for experimenting with creds, would like to update that to point to examples from here
Nate Otto: Glad to hear some good support for has achieved. Looking forward to working the action items. Action item could be: make recommendations of vocabularies where hasAchieved could be situated for good interoperability and community usage. Develop JSON-LD-schema/JSON-schema is a good action item