[Openid-specs-digital-credentials-protocols] 2023-11-23 SIOP/DCP meeting notes

[Joseph Heenan]

Spec Call Notes 2023-11-23

Attendees:

Kristina Yasuda
Joseph Heenan
Christian Bormann
Mark Haine
Torsten Lodderstedt
Giuseppe De Marco
Judith Kahrer
Paul Bastian
Bjorn Hjelm
Daniel Fett
Pedro Felix


Publishing a new revision of VCI

It’s a while since a numbered working group draft was published (Feb 2023). We should probably publish. Agreed to publish whatever is there at the end of today’s call.

There’s some inconsistencies in numbering. We last published 11, but the current draft is numbered 13. We agree to publish the latest as 12.


Pedro asked about timing for implementor’s draft.

Kristina proposed doing it once we’ve finished discussing the instances terminology. A lot of discussion. Torsten was in favour of starting implementor’s draft process immediately to lock in IPR protections. Others in favour of getting known breaking changes in. Torsten was worried that this could drift for a long time and expand in scope. Kristina said MS need the IPR protection, but not all the features they are implemented are in the draft yet. Christian suggested time boxing it. Joseph suggested getting the vote started before Christmas/New Year holidays, which gives us about 3 weeks to do any normative changes before starting WGLC and then the ID draft. Torsten agreed to that.


Issues/PRs


https://github.com/openid/OpenID4VCI/issues/89

Kristina asked why `proof_types_supported` isn’t the indication that a proof is required.

Torsten agreed but suggested that the “supported” suffix isn’t then applicable. Torsten suggested “required” but Joseph said that is awkward if there are multiple entries as only one is required. Torsten suggested just omitted “supported”, and if it was omitted then proof of possession is not required.

https://github.com/openid/OpenID4VCI/pull/70

Torsten has objections but they’re only editorial but has questions about the charging mechanism.

Joseph suggested we keep the recently raised charging issue separate and consider merging this for the original purpose of notifications.

David said there’s no difference of the credential failing at the last minute and the issuer doesn’t need this notification and this is adding complexity to all wallets.

Joseph said the complexity is low for wallets and we have issuers (Microsoft, Forgerock) telling us it’s valuable to them so the balance is towards merging.

Torsten suggested as we have implementors wanting it we should add it, and review in a few months if people are actually finding it useful.

David C agreed with that and will make a comment to that effect.

We agreed to keep the charging mechanism out of scope.


https://github.com/openid/OpenID4VCI/issues/91

Kristina explained her latest thinking from her most recent comment.

Torsten, Joseph & Judith agreed with latest thinking.

Daniel agreed to have a go at a terminology PR, and there may be follow ups needed.


https://github.com/openid/OpenID4VCI/pull/95

Pedro said he was okay to the response he got about lack of internationalisation.

Kristina asked about introducing pin_type to allow alphanumeric pins.

Agreed to add this. And to remove the 4-8 length restriction.

Consensus around renaming pin to tx_code before we merge.


https://github.com/openid/OpenID4VCI/issues/94

Pedro asked that we talk about this, but we didn’t have time today, but please review it and we’ll talk about it next week.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openid.net/pipermail/openid-specs-digital-credentials-protocols/attachments/20231123/6acfbd5a/attachment-0001.html>