The W3C Credentials Community Group

Verifiable Claims and Digital Verification

Go Back

Credentials CG Telecon

Minutes for 2020-03-03

Manu Sporny: Regrets+ manu
Error: (IRC nickname 'orie_' not recognized)[2020-03-03T17:09:44.670Z] <Orie_> i'd like to discuss next steps on did-key, webkms if possible :)
Will we have SDS WG - DIF/CCG discussion on today's call? thanks
Chris Winczewski: Scribe+ chriswinc
Chris Winczewski is scribing.
Chris Winczewski is scribing.
Scribe chriswinc
Jeff Orgel: Apologies I can not help in that fashion!

Topic: Introductions & Re-introductions

Topic: Announcements & Reminders

Christopher Allen: RWOT cancelled due to COVID-19, virtual hackathon offered as an option that week
KNOW Identity April 5 also cancelled
Christopher Allen: End of April is IIW
Chris Winczewski: KNOW Identity status is unknown
Christopher Allen: Action items
Dan Burnett: S/KNOW Identity April 5 also cancelled/ChristopherA: KNOW Identity April 5 also cancelled/
Error: (IRC nickname 'orie_' not recognized)[2020-03-03T17:21:09.733Z] <Orie_> happy to help move and close that stuff out :)
... Community group up for closure
... What do we need to do to close?
Dmitri Zagidulin: We moved all the relevant repositories
... redirects for needs to be setup
Joe Andrieu: Would be great to have an abandoned list
... people need an opportunity to fork them if necessary
Christopher Allen: If you are interested, please make sure you forward them into your own repos
... Digital verification homepage also needs updating
... Chairs will then formally close

Topic: Action items

... any questions regarding the closure? Attach them to issue 105
... Action item focus items are decided every Thursday. Send a note to the chairs if you want something prioritized.
... Work items page lists ongoing task forces as well as open items.
... Start with a report from the DID resolver task force which meets on Thursdays.

Topic: CCG Work Item Reports; DID Resolver Task Force

Kim Hamilton Duffy: ChistopherA: someone else from resolver task force to report?
Kim Hamilton Duffy: ChistopherA: so....Markus
Markus Sabadello: Haven't done much on spec content, dependencies on DID core spec [scribe assist by Kim Hamilton Duffy]
Kim Hamilton Duffy: ...Support for multiple reps of the did doc (abstract data model)
Kim Hamilton Duffy: ...Such items affect did resolution (e.g. which representation the resolver returns)
Kim Hamilton Duffy: ...Another such topic is matrix parameters
Kim Hamilton Duffy: ...A way of passing parameters
Kim Hamilton Duffy: ...Discussion over whether they should be removed in favor of query params
Kim Hamilton Duffy: ...Third major topic is metadata
Kim Hamilton Duffy: ...Types of metadata to support, different categories of metadata (specific to the did doc, specific to resolution process)
Kim Hamilton Duffy: ...E.g. "created" property
Kim Hamilton Duffy: ...What does it mean
Kim Hamilton Duffy: ...Not yet common agreement
Christopher Allen: One reason for this separate group was that it's not in scope of did-core [scribe assist by Kim Hamilton Duffy]
Kim Hamilton Duffy: ...But I've heard discussion that it may be brought into scope
Markus Sabadello: Some support that resolution "contract" would be brought into scope [scribe assist by Kim Hamilton Duffy]
Kim Hamilton Duffy: ...But implementation details would stay separate
Markus Sabadello: We will be discussing the interface to DID-core and matrix parameters in the upcoming meetings
Markus Sabadello: Open issue in the DID WG to discuss how much of DID Resolution should be shifted into DID Core:
Dmitri Zagidulin: DID Web, multiple DID Docs
Markus Sabadello: Open issue about metadata:
Markus Sabadello: Open issue about matrix parameters:
... e.g. can students host DID docs on an EDU hosted location? Is it legal in the syntax?
Hosting .well-known on non root paths?
Christopher Allen: DID WG requires W3C membership
... but the DID resolver task force only requires Community Group membership
Dmitri Zagidulin: Orie_ — no, more like, specifying that IF there is no path component, then use .well-known/, but if there IS a path component, use that _instead_ of .well-known.
...this is an opportunity to bring feedback
Markus Sabadello: DID Resolution weekly Thursday call:
Dmitri Zagidulin: So, it's like, can refer to a different did document than
Orie_: We have work items for DID Key and WebKMS
...moving repos from DigitalBazaar to central location. How do we do this to support broader work on these topics?
Christopher Allen: These are approved work items and the location of the work is up to the editors
...when it reaches certain maturity, it should be moved to CCG may be time to do this.
...are the leads of these work items present?
I am a code owner on webkms
...action added to add leads.
Error: (IRC nickname 'orie_' not recognized)[2020-03-03T17:43:39.395Z] <Orie_> but not on did:key :)
Dmitri Zagidulin: Will bring up subject with Manu and DLongley
Ok, I will tackle trying to move webkms along
ACTION: @dmitriz to ask DB to consider moving did:web to CCG
Christopher Allen: Any other questions for these work items? None.
...any other work items that would like to talk about status or ask for help?

Topic: other work items

Joe Andrieu: Some of these can move on e.g. DID use cases
Manu Sporny: (From IRC) - happy to move WebKMS and did:key to CCG. [scribe assist by Manu Sporny]
...Use cases has been formally handed over. Document is out of date.
Christopher Allen: Steps to close this out include changing references, archiving
ACTION: @joeandrieu will puzzle DID Use Cases how to wrap it
...Can JoeA wrap this up?
Joe Andrieu: Yes, with help for steps
Christopher Allen: This is a success point when a work item is taken up by a formal WG or other standards body. We should celebrate.
Joe Andrieu: The other one was the DID Explainer
ACTION: @burn & @JoeAndrieu to close out explainer work item
...the work has move to a different place and a pointer is needed to the new place
Christopher Allen: Any feedback from the community regarding other open work items?
...inactive items are subject to closure or archive
Orie_: been discussion on a DID key issue
...and matrix parameters
...the problem is that DID Docs don't have all the information they need at initialization
...updates however are expensive in time or cost
...matrix parameters offer to extend original DID doc if they are signed
...would allow you to sign an attribute e.g. service endpoint
Christopher Allen: Related to a BTCR item
...we are stalled on a non JS-based signing method DID key, begin with a key pair
...put on BTC and until the transaction is confirmed we don't have a DID
Yes, many methods that rely on ledger anchoring suffer from these registration timing issues.
...need the block ID of transaction
...all you have in the interim is a key
...need to go to extensions to get override key needs to do everything or at least delegation
Error: (IRC nickname 'orie_' not recognized)[2020-03-03T17:57:10.978Z] <Orie_> here is mine for example:
...when DID doc is extended, that document will modify key authority for example
...we don't quite know how to do that might get different DID extensions based who you are and what your rights are
Joe Andrieu: @Orie_ that is an oddly formed DDO
...were hoping to dive into this at RWOT but need a backup plan now
Christopher Allen: @Dmitriz
Yes, opportunity for alignment accross methods wrt timing and property issues.
Dmitri Zagidulin: What is being proposed is a way to extend the trust model pass in essentially a JSON patch
Christopher Allen: Are there an example of json patch mechanism? any urls about this? resolver - please also add or remove
...may resolve issues in many DID methods
Christopher Allen: BTCR method thought this would reside in the verifiable presentation
Dimitriz: good use case, add it to issue #5
Jonathan Holt: Is the concern derived keys?
Christopher Allen: Option to pass an xpub instead of a key
...or maybe even every time you sign
...may have to prove capability to get DID doc
...could be cryptographic. potentially solves phoning home
Christopher Allen: @JoeAndrieu
...BTCR focused on privacy, security without using selective disclosure
Joe Andrieu: DID doc identifies that there are more parameters that should be checked
...initial state use case.
Sidetree for example has initial state gurantees
...third - augment with JSON patch. This seems to be an anti-pattern
...feels broken as it is not deterministic
Orie_: agree with the three bucket separation
... method itself is responsible for returning the document
Christopher Allen: (Closing queue)
...up to the method maintainers how to do this
...not a big difference with augmenting doc as it is has the same effect as losing the key
Dmitri Zagidulin: Three categories are equivalent
...discussion should continue
Joe Andrieu: Multiple versions could be signed...
Joe Andrieu: Gotta go. thanks, all.
Christopher Allen: Offers to help with DID extensions if you provide URL