See below..

Phil
[email protected]




On 2010-12-15, at 4:09 PM, Manger, James H wrote:

> Ø  What is the reasoning behind the lack of a client_id parameter in requests 
> to protected resources?
>  
> When the client app is acting on its own behalf (with the app’s own long-term 
> credentials)… the client_id is included as part of authenticating the client 
> app (as a query/form parameter, or as the user-id field in a “Authorization: 
> BASIC” header, or in any other authentication mechanism).
>  
> When the client app is acting on behalf of a user (with credentials from an 
> OAuth token response)… the client_id is NOT included. The credentials from 
> the token response are sufficient. Adding client_id here is unnecessary (the 
> server can include it in the token if it is convenient for protected 
> resources), and harmful (it means the protocol that uses the credentials from 
> the token response cannot look like a normal authentication protocol).

It is useful for the resource authorization policy server to know what client 
is acting on behalf of what user.  E.g. if a financial aggregator was using 
OAuth to access a bank, the bank systems can differentiate rights between 
normal end-user access and a client app access (e.g. read-only regardless of 
scope).  I know scope already covers some of this, but it seems like an 
important distinction if for some reason a policy server needs to apply 
additional rules for clients beyond those of scope.
>  
> --
> James Manger
> _______________________________________________
> OAuth mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/oauth

_______________________________________________
OAuth mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/oauth

Reply via email to