[Openid-specs-ab] Spec Call Notes 7-May-20
Mike Jones
Michael.Jones at microsoft.com
Thu May 7 15:08:15 UTC 2020
Spec Call Notes 7-May-20
Mike Jones
Brian Campbell
Roland Hedberg
Tim Cappalli
Bjorn Hjelm
George Fletcher
Bhupinder Singh
John Bradley
Federation Specification
We released a new draft last week
There was discussion on the list of using signed requests rather than private_key_jwt
Roland is working on that change
We'd planned an interop event at TNC in June in Bristol
We're planning on doing that virtually instead
Roland is deploying a federation that testers will be able to use
We know of three implementations at present
Roland plans to do the signed request update before the interop
Identiverse
We discussed the ways that Identiverse may be run
Mike was hoping to get interactive feedback on the Federation work
OAuth JAR
Brian did a PR to address the PAR issue
There's another PR by Torsten using metadata to determine algorithms
https://bitbucket.org/Nat/oauth-jwsreq/pull-requests/4/
That needs to be merged
Nat and John then need to publish an updated draft again
Safari Bounce Tracking Proposal
George described Apple's bounce tracking proposal
See https://github.com/privacycg/proposals/issues/6 and
https://github.com/privacycg/meetings/blob/master/2020/telcons/04-23-bouncetracking-minutes.md
This is in a Safari tech preview release
Federation redirects look like this
This is another possible set of browser changes that could affect identity flows
Like the SameSite cookie changes
This may interact with the IsLoggedIn proposal https://github.com/WebKit/explainers/tree/master/IsLoggedIn
and the Google WebID proposal https://github.com/samuelgoto/WebID
Open Issues
https://bitbucket.org/openid/connect/issues?status=new&status=open
#1164-#1166 Federation issues now assigned to Roland
Roland said that there's some other issues that he's addressed that he should close
#1160 Registration 2 - Should data: URLs be allowed as valid logo_uri values?
No new thoughts during the call
#1161 Key rotation should require a delay between publishing a key and starting to use it?
There's been more discussion on the issue, including about what the certification code is doing
There's no evidence that these possible attacks have ever occurred in practice
Roland said that in SAML federations, new keys are typically added days before use
But in some circumstances, keys have to be changed faster than that
#1086 Core 5.6.2 - chaining Distributed Claims
We should investigate this
The question is whether recursion is allowed
There doesn't appear to be a reason why this shouldn't work
Next Call
The next working group call is Monday, May 11 at 4pm Pacific Time
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openid.net/pipermail/openid-specs-ab/attachments/20200507/d6c48bcd/attachment.html>
More information about the Openid-specs-ab
mailing list