Wiki

Clone wiki

fapi / FAPI_Meeting_Notes_2023-03-15_Atlantic

FAPI WG Agenda & Meeting Notes (2023-03-15)

The meeting was called to order at 14:02 UTC.

1.   Roll Call (Dave/Nat)

  • Attending: Nat, Gail, Joseph, Mike, Tim, Aaron, Brian, Daniel, Dave, Filip, Geroge, Justin, Marcus, Kosuke, Pedram, Kelley, Pieter, Dima
  • Regrets: Chris
  • Guest:

2.   Adoption of Agenda (Dave/Nat)

  • Adopted as presented as draft agenda.

3.   Mini-kick off of FAPI2 Workpackage 2 project (Gail/Marcus)

  • Scope is FAPI CIBA, Dynamic Client Registration (including Australian variant) and FAPI 2.0 Message Signing (including JARM, JAR. jIR, HttpSiG)
  • Meeting at 8 AM CET.

4.   Events (Mike L)

  • Workshop on 1 PM Apr. 17. Details to be published this week.

6.   External Orgs & Liaisons (Mike L.)

6.1.   Brazil

  • No material updates.
  • Certification request continues to come in.

6.2.   Saudi

  • KSA FAPI Profile Framework is not publicly available yet. Quality control is being performed.
  • This caused problems for parties that wanted to comply with the profile.
  • Good news is that it seems it is possible to publish FAPI-specific components at openid.net. It will be confirmed shortly.

7.   Draft Updates

7.1.   FAPI 2 Msg Signing (Dave)

  • Package was sent to the foundation secretary.
  • Question was raised whether the title should have "Draft" or "Implementer's draft"
    • It should be "Draft" because it has not been voted on.
  • The draft will be updated and sent to the foundation secretary again.

7.2.   All other drafts: Put "draft" in the title (Nat)

  • Editors, please make sure that your draft has "draft" or "Implementers draft #" in the title.
  • Also, please make sure to put warning text.

8.   PRs (Dave)

  • Apart from one PR that we are parking until HTTP signature is settled, there is no standing PR.

9.   Issues (Dave)

9.1.   FAPI2SP: Note about client assertion audience looks misleading

9.2.   FAPI2SP appears to permit response_types "id_token", "id_token token" and "none"

  • #577
  • https://bitbucket.org/openid/fapi/issues/577/fapi2sp-appears-to-permit-response_types
  • Brian's wording was discussed.
  • It was pointed out that while it is rejecting the response types quoted, it does not others.
  • Adding parameters (esp. tokens) would create a new authentication protocol and nullify the security analysis.
  • In view of this, it was suggested to lock it down to response_type=code while making it conditional that future extensions, such as CIBA, can be made. (They need a fresh security analysis and we are doing that in FAPI2 Workpackate 2 sponsored by the AU government.)
  • Filip came up with wording that sounded reasonable. He will put it in this ticket.

10.   AOB (Nat)

  • none

The call adjourned at 14:59

Updated