[Openid-specs-ab] SIOP Special Topic Call Notes 20-Jul-23
Michael Jones
michael_b_jones at hotmail.com
Thu Jul 20 15:24:43 UTC 2023
SIOP Special Topic Call Notes 20-Jul-23
Mike Jones
Kristina Yasuda
Mike Leszcz
Felix Linkler
Brian Campbell
Bjorn Hjelm
David Waite
Dmitri Zagidulin
Joseph Heenan
Mark Dobrinic
Nander Stabel
Pedro Felix
Oliver Terbu
IETF
The deadline to upload OAuth slides is Friday
https://datatracker.ietf.org/meeting/117/session/oauth
OSW
https://oauth.secworkshop.events/osw2023
Registrations are open
Digital Credentials Protocols Working Group
The new Digital Credentials Protocols working group was approved
There isn't a deadline to sign a new Contribution Agreement
One is needed if your existing agreement doesn't specify the "all working groups" option
We likely need to continue work on some specs in the Connect working group for a bit longer
Kristina observed that we may need to create Implementer's Drafts to transition some of the specs to the new WG
Merging Policy
Kristina said that PRs will be merged once there are approvals from two editors and sufficient time to review
Taka had requested that we not spend time on the calls discussing editorial PRs
Pull Requests
https://bitbucket.org/openid/connect/pull-requests/
PR #542: adding key proof verification steps
Needs additional reviews
PR #551: feat: [OpenID4VCI] added trust_chain in proof types
There are enough approvals to merge
PR #557: Adds encrypted credential response
Oliver believes changes need to be made based on Kristina's comment
Change to have the issuer request encrypted responses
Requested encryption "alg" and "enc" have been added to the issuer metadata
Brian thought it would be weird to also have a require_encryption metadata parameter
Brian thought that using jwks is overkill - only one key is needed
Mike observed that you can tell what algorithms the recipient supports through metadata
PR #577: add security considerations on TLS (Issue #1621)
We will reference the TLS BCP
PR #455: OID4VP add train client id scheme
Needs additional reviews
PR #570: clarify requirements when credential offer is not signed (issue #1687)
Clarifications about whether to use alg:none are needed
PR #360: Add an access token hash to the proof of possession
We plan to close this unless Richard says why not to
PR #559: Change last_name to family_name
Ready for merging
PR #560: Change driver's license to driving license
Ready for merging
PR #561: Tighten Introduction
Needs another editor approval to merge
Findings on BLE
Felix Linkler told us about his OpenID4VP over BLE findings
Used the Tamarin prover to analyze the protocol
Abstracted protocol to enable modelling using a formal model
One property is Secrecy
Another property is Injective Agreement
The design should meet certain security requirements against a certain adversary
Specify both!
Felix will file issues requesting additions/changes
Kristina observed that we will want to work on this spec in the new working group
But it's fine to file issues in the Connect working group in the meantime
Issues
https://bitbucket.org/openid/connect/issues?status=new&status=open&component=SIOP&component=Verifiable%20Presentation&component=Credential%20Issuance
#1923: OID4VCI: Enable issuance of the same credential type
Kristina asked if it would be useful to add identifiers
If so, also add to Credential Offer
The identifiers would be managed by the issuer
Pedro Felix said that their meaning would be opaque to the Wallet
Kristina said that when the Wallet receives two identifiers, it should know to send two credential requests
Kristina talked about ways that the identifiers might not be opaque
More reviews are requested
An example about multiple birth certificates for multiple children was discussed
Next Call
The next call would be Monday, July 24th at 4pm Pacific Time but this conflicts with IETF
The call may be cancelled
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openid.net/pipermail/openid-specs-ab/attachments/20230720/51d2601f/attachment-0001.html>
More information about the Openid-specs-ab
mailing list