[OpenID] OpenID schemas

[Andrew Arnott]

BTW, one mitigation in the meantime to RPs not having to send attribute
requests as sreg plus the 3 AX type URI formats
<http://blog.nerdbank.net/2009/03/how-to-pretty-much-guarantee-that-you.html>that
are seen in the wild is for OPs to list the supported extensions and (if AX
applies) the AX type URIs supported in their users' identity pages' XRDS
document and on their OP Identifier's XRDS document.

For instance, an RP performing discovery on some identifier might see sreg's
extension URI in the OPs XRDS doc and just send sreg.  For another OP it
might see AX attribute Type URIs in the axschema.org format and know that it
should send an AX request in that format, and in another it might see
schema.openid.net and use that format.

If OPs would publish the extensions they support and the AX attributes they
offer, then RPs can discover this and send the minimal request necessary --
only sending all 4 formats of attribute requests to OPs that don't advertise
which they support.

While this might sound like a lot of work at the RP to special case all the
attribute requests based on discovery, this can be done in the libraries so
that RPs don't have to individually write it all up.  Besides, it's better
than RPs special casing for individual OPs by OP endpoint as they are doing
today.

--
Andrew Arnott
"I [may] not agree with what you have to say, but I'll defend to the death
your right to say it." - S. G. Tallentyre


2009/6/18 John Bradley <john.bradley at wingaa.com>

> The first thing is the Board or whomever deciding that the OIDF is going to
> run a schema registry for AX.
>
> If they don't want to then it is a different problem.
>
> The we need to establish a standard set of URI that map to SREG.
>
> We could start with the process defined in
> http://www.axschema.org/specs/openid-attribute-types-1_0-02.html
>
> The doc needs to change schema.openid.net to schemas.openid.net
>
> We need top decide if we want to define types as well.
> http://www.axschema.org/specs/identity-attribute-metadata-1_0-01.html
>
> I think we could adopt the 9 common URI almost immediately
> http://schemas.openid.net/contact/email/namePerson/friendly
> Alias/Username        openid.sreg.nickname
> http://schemas.openid.net/contact/email
>       Email                   openid.sreg.email
> http://schemas.openid.net/contact/email/namePerson
>  Full name               openid.sreg.fullname
> http://schemas.openid.net/contact/email/birthDate
> Birth date              openid.sreg.dob
> http://schemas.openid.net/contact/emailg/person/gender          Gender
>              openid.sreg.gender
> http://schemas.openid.net/contact/email/contact/postalCode/home Postal
> code    openid.sreg.postcode
> http://schemas.openid.net/contact/email/contact/country/home    Country
>               openid.sreg.country
> http://schemas.openid.net/contact/email/pref/language           Language
>              openid.sreg.language
> http://schemas.openid.net/contact/email/pref/timezone           Time zone
>               openid.sreg.timezone
>
> The process for registering new URI is very similar to the one that the
>  ICF uses for claim types.
>
> We do need to check that all of the Sxip documents are covered under there
> contribution agreement.
>
> The important thing is to officially define the above table so that we have
> one common set of URI and not two as we do now.
>
> The longer term schema and registration process could be a WG or a Board
> committee of some sort.
>
> The administrative process of running the registry doesn't really have IPR
> so could be separate from the WG.
>
> I pointed out the problem during the OSIS interop testing.   The state of
> AX interop is embarrassing.
>
> Would Google support a standard set of AX attribute URI if they are
> defined.
>
> Currently you are using the original Sxip ones at axshema.org.
>
> Regards
> John B.
>
>
> On 17-Jun-09, at 7:18 PM, Breno de Medeiros wrote:
>
>  Agreed. What work needs to be done today?
>>
>> On Wed, Jun 17, 2009 at 4:13 PM, John Bradley<john.bradley at wingaa.com>
>> wrote:
>>
>>> Leave the other parts of AX 2.0 like defining structured attributes etc
>>> to
>>> the AX 2.0 work group if it gets going.
>>>
>>> We need a standard set of attributes for AX 1.0 now,  not in a year.
>>>
>>> John B.
>>> On 17-Jun-09, at 6:57 PM, Breno de Medeiros wrote:
>>>
>>>  Context failure: Too many unresolvable acronyms.
>>>>
>>>> On Wed, Jun 17, 2009 at 3:35 PM, John Bradley<john.bradley at wingaa.com>
>>>> wrote:
>>>>
>>>>>
>>>>> Leave the other bits to the AX 2.0 WG.
>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> --Breno
>>>>
>>>> +1 (650) 214-1007 desk
>>>> +1 (408) 212-0135 (Grand Central)
>>>> MTV-41-3 : 383-A
>>>> PST (GMT-8) / PDT(GMT-7)
>>>>
>>>
>>>
>>>
>>
>>
>> --
>> --Breno
>>
>> +1 (650) 214-1007 desk
>> +1 (408) 212-0135 (Grand Central)
>> MTV-41-3 : 383-A
>> PST (GMT-8) / PDT(GMT-7)
>>
>
> _______________________________________________
> general mailing list
> general at openid.net
> http://openid.net/mailman/listinfo/general
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openid.net/pipermail/openid-general/attachments/20090619/21bc2bda/attachment.htm>