One example is when used as a signed request to the authorization server, as is
done in http://tools.ietf.org/html/draft-sakimura-oauth-requrl-05.
-- Mike
-----Original Message-----
From: Hannes Tschofenig [mailto:[email protected]]
Sent: Monday, July 14, 2014 10:45 AM
To: Mike Jones; Brian Campbell; John Bradley
Cc: [email protected]
Subject: Re: [OAUTH-WG] Dynamic Client Registration: jwks / jwks_uri
Hi Mike,
sticking with working group document is fine.
However, the first example does not make sense to me.
[maybe my brain is a bit empty at the moment]
When is a JWT signed by the client and then sent to the Authorization Server
other than in the Assertion draft that I mention in the second example?
Ciao
Hannes
On 07/14/2014 06:16 PM, Mike Jones wrote:
> I'd rather that we stayed with working group drafts in the examples.
> So I would counter-propose the following text:
>
> "The public key(s) referenced by "jwks_uri" (or contained in the
> "jwks") can be used in a variety of use cases. For example, the
> signature of a JWT [I-D.ietf-json-web-token] signed by the client can
> be verified by the authorization server using these keys. Another
> example is that the authorization server can use the indicated public
> keys to verify a request to the token endpoint that utilizes the JWT
> assertion profile as described in Section 4.2 of
> [I-D.ietf-oauth-assertions]."
>
> -- Mike
>
> -----Original Message----- From: OAuth [mailto:[email protected]]
> On Behalf Of Hannes Tschofenig Sent:
> Monday, July 14, 2014 2:42 AM To: Brian Campbell; John Bradley Cc:
> [email protected] Subject: Re: [OAUTH-WG] Dynamic Client Registration:
> jwks / jwks_uri
>
> What about the following text:
>
> jwks_uri
>
> .... <previous text in Section 2 of
> http://tools.ietf.org/html/draft-ietf-oauth-dyn-reg-18> .....
>
> "The public key(s) referenced by jwks_uri (or contained in the jwks)
> can be used in a variety of use cases. For example, the AS can use the
> indicated public key to verify a request to the token endpoint that
> utilizes the JWT assertion profile as described in Section 4.2 of
> [I-D.ietf-oauth-assertions]. Another use case is for the AS to use the
> public key of a client to encrypt a symmetric proof-of-possession key
> sent to the client, as described in Section 4.2 of
> [I-D.bradley-oauth-pop-key-distribution]."
>
>
> Ciao Hannes
>
>
>
>
>
>
>
> On 07/08/2014 09:43 PM, Brian Campbell wrote:
>> +1 to John's #3. The others could maybe be described in somewhat
>> abstract terms as examples of those "higher level protocols that use
>> signing or encryption."
>>
>> On Tue, Jul 8, 2014 at 12:33 PM, John Bradley <[email protected]>
>> wrote:
>>> In Connect these public keys are used to: 1 verify the signature of
>>> request objects (Signed Requests), something not in OAuth yet, and
>>> part of what the description calls higher level protocols. 2 encrypt
>>> the responses from the user_info endpoint or id_token (also not part
>>> of OAuth directly at this point)
>>>
>>> 3 validate requests to the token endpoint authenticated by the JWT
>>> assertion profile I think this is legitimate OAuth use.
>>>
>>> Whew for the PoP specs: 4 used to encrypt the symmetric proof key in
>>> a JWK sent to the client
>>> http://tools.ietf.org/html/draft-bradley-oauth-pop-key-distribution-
>>> 0
>>>
>>>
1#page-7
>>> 5 used to provide a PoP key for the client to the AS as part of
>>> registration rather than passing the JWK on each request to the
>>> token endpoint.
>>>
>>> So the keys in the JWK can be used a number of ways by the AS.
>>>
>>> I think we could reference 3 and 4 as examples to be safe.
>>>
>>> John B.
>>>
>>>
>>> On Jul 8, 2014, at 3:04 PM, Mike Jones <[email protected]>
>>> wrote:
>>>
>>>> Was there specific language that had been discussed to be added for
>>>> this? If not, could someone please create some?
>>>>
>>>> Thanks, -- Mike
>>>>
>>>> -----Original Message----- From: OAuth
>>>> [mailto:[email protected]] On Behalf Of Hannes Tschofenig
>>>> Sent: Tuesday, July 08, 2014 5:09 AM To: [email protected]
>>>> Subject: [OAUTH-WG] Dynamic Client Registration: jwks / jwks_uri
>>>>
>>>> Hi all,
>>>>
>>>> in my earlier review I had noted that the semantic of the fields is
>>>> underspecified, i.e., it is not clear what these fields are used
>>>> for.
>>>>
>>>> In private conversations I was told that an informal reference to a
>>>> potential use case will be added. I don't see such reference with
>>>> version -18.
>>>>
>>>> Ciao Hannes
>>>>
>>>> _______________________________________________ OAuth mailing list
>>>> [email protected] https://www.ietf.org/mailman/listinfo/oauth
>>>
>>> _______________________________________________ OAuth mailing list
>>> [email protected] https://www.ietf.org/mailman/listinfo/oauth
>>
>> _______________________________________________ OAuth mailing list
>> [email protected] https://www.ietf.org/mailman/listinfo/oauth
>>
>
_______________________________________________
OAuth mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/oauth
