- From: James Hawkins <jhawkins@chromium.org>
- Date: Tue, 8 May 2012 15:15:44 -0700
- To: Greg Billock <gbillock@google.com>
- Cc: Charles Pritchard <chuck@jumis.com>, Eiji Kitamura <agektmr@google.com>, Doug Schepers <schepers@w3.org>, public-web-intents@w3.org, Harry Halpin <hhalpin@w3.org>, nov matake <matake@gmail.com>
- Message-ID: <CAO800SyB5wOepWQgj0jyWVExByjOVQKzU0fzqXcPwM_WnrEh1w@mail.gmail.com>
Greg, let's pull discussions about oauth into a separate thread, since the use cases for oauth, i.e. resource authentication, and openid, i.e. identification, are not the same. I think the former is important to solve as well, but they're both tricky so we should separate the discussions. Thanks, James On Tue, May 8, 2012 at 2:11 PM, Greg Billock <gbillock@google.com> wrote: > The OAuth implicit grant flow [1] looks pretty amenable to a web > intents adaptation. It would map to an explicit intent -- the client > is looking for an access token for a particular resource, capability, > or permission. Being able to capture that via a web intent as opposed > to a redirect flow is interesting from a user perspective for being > able to take advantages of the web intents UI transitions -- an > overlapped UI surface which is under the control of the provider and > can be properly attributed. > > On the client side, the ability to have this act as an async call, > which doesn't need to tear down and rebuild all the desired app state > while the auth call completes, is pretty attractive. > > > [1] http://tools.ietf.org/html/draft-ietf-oauth-v2-25#section-4.2 > > On Tue, May 8, 2012 at 12:51 PM, Charles Pritchard <chuck@jumis.com> > wrote: > > On 5/8/2012 10:44 AM, James Hawkins wrote: > >> > >> As far as the namespace is concerned, I'm not entirely sure we can't > have > >> a high-level 'identify', but I do believe we should make use of existing > >> technologies instead of inventing a new protocol. Each identity > protocol > >> would have its own action, e.g. http://specs.openid.net/auth/2.0 for > OpenID. > >> The reason for this is that the client needs to verify the > authenticity of > >> the identity w/ the server, which requires messaging beyond the simple > >> input/response provided by the basic Web Intents layer. > > > > > > Yeah my take on it is quite similar. I did the OAuth route in our web > apps; > > it doesn't have a standard identity, though every provider has something > > like a /self/me end-point which returns JSON describing the user. > > It seems to me that I'd be passing back: {token:'arbitrary', 'endpoint': > > 'https://url'}; > > > > From there, I can do client-side via <iframe> and web messaging for a > > channel, or server-side for identity. > > > > -Charles > > > > > > >
Received on Tuesday, 8 May 2012 22:16:46 UTC