[OpenID] OpenID 2.1 Identifier Types --> WAS [Discovery

[John Bradley]

I have talked to Luke at facebook.

There is no grand conspiracy against XRI, or plot not to conform to  
the spec.

They like many of us are resource constrained, and are incrementally  
rolling out features.   Some with novel twists at that:)

I believe the library they are using supports XRI resolution, so  
saying that it is too complicated is a bit of an overstatement.

If anything people on this list have told them that XRI is being  
removed from the 2.1 spec so don't implement it.

Those same people point to them not implementing it as grounds to  
remove it from the spec:)

It is a bit tiresome.

I expect Facebook will implement the spec in there own time.

Once the new XRD 1.0 spec is out XRI 3.0 will become no more  
complicated to integrate than any other identifier that uses XRD 1.0  
based discovery.

We do however have issues around identifiers in the core spec that  
will need to be addressed.

Are the identifier abstractions that we are exposing sensible across  
multipe identifier types or tailored specificly to URI?

At the moment we have openid.claimed_id that is the only authoritative  
identifier,  but may have no clear meaning or relationship to the user  
input.

The openid.identity  might not be global in scope it could be a  
localID from the XRD, delegate from a rel tag or some normalized  
version of the user input.

We need to do a better job at abstracting the authentication/discovery  
part of openID from the API we are presenting to applications.

Without a clear understanding of that abstraction adding other  
identifier types beyond URI will be a challenge from a UI perspective.

John B.

On 4-Jun-09, at 9:59 PM, general-request at openid.net wrote:

> Date: Thu, 4 Jun 2009 18:42:34 -0700
> From: Peter Williams <pwilliams at rapattoni.com>
> Subject: Re: [OpenID] OpenID 2.1 Identifier Types --> WAS [Discovery
> 	for Email like identifiers]
> To: Chris Messina <chris.messina at gmail.com>, "sappenin at gmail.com"
> 	<sappenin at gmail.com>
> Cc: Santosh Rajan <santrajan at gmail.com>,	"general at openid.net"
> 	<general at openid.net>
> Message-ID:
> 	<BFBC0F17A99938458360C863B716FE46398DCE8FA8 at simmbox01.rapnt.com>
> Content-Type: text/plain; charset="Windows-1252"
>
> Facebook are non conforming - if they didn't allow for user's to use  
> their hxris. Shame on them. Its not exactly hard.
>
> Using openid mechanisms to reimplement facebook connect (with  
> variant messages) is a both a move forward and a move back. If  
> openid is just a tech library rather than a movement for uci, im not  
> sure it needs the overhead of the foundation.
>
> ________________________________