OIDC Deployment Working Group

[Eric Goodman <>]

I separately sent a note to Nathan that I'm on the agenda for the UC-wide API 
"standards" workgroup to discuss what people are doing/looking with 
OIDC/OAuth. That meeting is a week from Friday, so hopefully I'll have some 
data points to provide after that meeting -- which is unfortunately after our 
next OIDC group meeting. Vaguely related, I'm at a conference next Tuesday, 
so may miss next week's call.

--- Eric

-----Original Message-----
From: 

 
<>
 On Behalf Of Steven Carmody
Sent: Thursday, August 9, 2018 8:43 AM
To: 

Subject: a start, OAuth2 use cases, thoughts on addressing each

Hi,

On last week's call, Alan, Roland, and I volunteered to identify the 
most common OAuth2 use cases currently in place on campuses; step 2 was 
to identify best practice for addressing each; step 3 would be to see 
how much of the required support is already available in the 
Shibboleth-GEANT-Finish OIDC work.

We had an email conversation, but it quickly became clear that this 
would benefit from participation by the broader group.

I've created a new google doc:

https://docs.google.com/document/d/15G_YRWisEa-5Kj5gU_D-i4MLMh-IJWl0Q5CAsqSy4ag/edit?usp=sharing

everyone should be able to edit that doc. It includes large portions of 
the email that Alan and I exchanged.

I'm on the way out the door to a family wedding (this won't be anywhere 
near as much fun as our daughter's wedding in Dublin this past June -- 
that was five days of parties!), but fun nonetheless !

As you'll quickly see, this doc needs a lot of editing. More than that, 
it needs people with more OAuth2 knowledge than me working on it. I'm 
also not completely sure that I may be operating with out-of-date 
assumptions. For instance, a couple of years ago most Native Apps were 
using OAuth2 to authenticate users -- is that still true ? The current 
recommendation for Native Apps seems to be to use "Authorization Code 
Grant Type with PKCE (Proof Key for Code Exchange)" -- do apps really 
use that ?

The doc also leaves too many open questions hanging ... again, here's 
hoping that people with more knowledge and more current knowledge can 
improve the doc.

Here's hoping people have a chance to review this, mark it up, improve 
it, and perhaps discuss it on the next call. Unfortunately, I'll miss 
the next call. Sorry.