While Eve is right in principle, I think it's better for everyone if
things remain consistent across different endpoints where possible. This
is why registration now does form input and json output, for instance.
Still simple, not as RESTy, but ultimately REST-friendly, which is all
OAuth ever sought to be.
-- Justin
On 01/23/2013 03:12 PM, Eve Maler wrote:
I'd say OAuth is an HTTP security mechanism or protocol (50,000-foot view). To
the extent that it, or its constituent parts, defines endpoints, it also is a
security API or set of APIs (10,000-foot view). Since its endpoints use HTTP,
this gives the opportunity to ask the question: how RESTful should its APIs be?
I'm hearing one firm design principle:
- Make new constituent parts be consonant with the rest of OAuth design
And a couple of competing soft design principles:
- Make it easy to develop endpoints (especially but not exclusively client-side)
- Make it flexible to accommodate lots of situations
I'm suggesting considering another one, possibly ranking it lower than others:
- Make its endpoints operate in a RESTful, resource-oriented way
(While OAuth hasn't gone through this exercise, UMA has, and we included a DP
about RESTfulness; you can see the results here:
http://kantarainitiative.org/confluence/display/uma/UMA+Requirements That's why
we defined draft-hardjono-oauth-resource-reg the way we did.)
The dyn reg spec is a kind of bootstrapping spec, so one can expect that it
would be out of the REST mainstream in any case -- its credential provisioning
function is sui generis. But the functions to create and update metadata look
like pretty ordinary CRUD functions that usually correspond to various POST or
PUT patterns in REST. If the client metadata were conceived of as a true web
resource, it's not wildly out of left field to consider defining their creation
and management in high-REST-maturity ways, and even imagine what (say) DELETE,
PATCH, and GET might mean, even if not allowed in the spec today.
(Justin, I promise I'm not trying to give you a hard time. :-)
Eve
On 23 Jan 2013, at 10:54 AM, Phil Hunt <phil.h...@oracle.com> wrote:
My understanding is OAuth is an HTTP protocol. It is not intended to be REST
specific or by implication be RESTful.
@independentid
www.independentid.com
phil.h...@oracle.com
On 2013-01-23, at 10:40 AM, Todd W Lainhart wrote:
On the other hand, it's a useful exercise to imagine how much more benefit could
potentially be gotten "for free" if we look at it through a pure-REST lens, not
just with what's already been specified but the whole picture.
+1
-- Todd
From: Eve Maler <e...@xmlgrrl.com>
To: Sergey Beryozkin <sberyoz...@gmail.com>,
Cc: Paul Bryan <em...@pbryan.net>, "oauth@ietf.org WG" <oauth@ietf.org>
Date: 01/23/2013 12:18 PM
Subject: Re: [OAUTH-WG] Concerning OAuth introspection
Sent by: oauth-boun...@ietf.org
Agreed that REST purity may come at a cost that's too high. On the other hand, it's a
useful exercise to imagine how much more benefit could potentially be gotten "for
free" if we look at it through a pure-REST lens, not just with what's already been
specified but the whole picture.
If what you're registering is a client descriptor, then creating a new one,
updating an existing one, deleting, and even patching could come for free if
something like the following framework is used:
http://tools.ietf.org/html/draft-pbryan-http-json-resource-03
With standard libraries possibly floating around to support this framework (I
think Paul B wrote one; maybe he open-sourced it?), it starts to become a lot
cheaper to support client registration on both sides of the interaction.
Eve
On 23 Jan 2013, at 8:34 AM, Sergey Beryozkin <sberyoz...@gmail.com> wrote:
On 23/01/13 15:47, Justin Richer wrote:
Which brings up an interesting question for the Registration doc: right
now, it's set up as a single endpoint with three operations. We could
instead define three endpoints for the different operations.
I've not been keen to make that deep of a cutting change to it, but it
would certainly be cleaner and more RESTful API design. What do others
think?
IMHO the purity should be balanced against the practicality/simplicity
of the implementation.
Talking about 3 endpoints at the spec level may be treated as the exact
requirement to have 3 separate application endpoints for the single type
of activity, the registration. Can the spec be re-worded such that
"resources" are used instead of endpoints or similar, example, "resource
available at /a will support the following, at /b - something else", or
may be something similar, thus it will read better too from the design
point of view, and let implementers to use 1 endpoint or 3 ones,
whichever way they prefer it
Thanks, Sergey
-- Justin
On 01/22/2013 08:05 PM, Nat Sakimura wrote:
"Action" goes against REST principle.
I do not think it is a good idea.
=nat via iPhone
Jan 23, 2013 4:00?Justin Richer<jric...@mitre.org> ??????:
(CC'ing the working group)
I'm not sure what the "action/operation" flag would accomplish. The idea behind having different
endpoints in OAuth is that they each do different kinds of things. The only "action/operation" that
I had envisioned for the introspection endpoint is introspection itself: "I have a token, what does it
mean?"
Note that client_id and client_secret *can* already be used at this endpoint if
the server supports that as part of their client credentials setup. The
examples use HTTP Basic with client id and secret right now. Basically, the
client can authenticate however it wants, including any of the methods that
OAuth2 allows on the token endpoint. It could also authenticate with an access
token. At least, that's the intent of the introspection draft -- if that's
unclear, I'd be happy to accept suggested changes to clarify this text.
-- Justin
On 01/22/2013 01:00 PM, Shiu Fun Poon wrote:
Justin,
This spec is looking good..
One thing I would like to recommend is to add "action"/"operation" to the
request. (and potentially add client_id and client_secret)
So the request will be like :
token REQUIRED
operation (wording to be determine) OPTIONAL inquire (default) | revoke ...
resource_id OPTIONAL
client_id OPTIONAL
client_secret OPTIONAL
And for the OAuth client information, it should be an optional parameter (in
case it is a public client or client is authenticated with SSL mutual
authentication).
Please consider.
ShiuFun
Eve Maler http://www.xmlgrrl.com/blog
+1 425 345 6756 http://www.twitter.com/xmlgrrl
_______________________________________________
OAuth mailing list
OAuth@ietf.org
https://www.ietf.org/mailman/listinfo/oauth
_______________________________________________
OAuth mailing list
OAuth@ietf.org
https://www.ietf.org/mailman/listinfo/oauth
Eve Maler http://www.xmlgrrl.com/blog
+1 425 345 6756 http://www.twitter.com/xmlgrrl
_______________________________________________
OAuth mailing list
OAuth@ietf.org
https://www.ietf.org/mailman/listinfo/oauth
_______________________________________________
OAuth mailing list
OAuth@ietf.org
https://www.ietf.org/mailman/listinfo/oauth
