Went Through it With Brad

[Johannes Ernst]

The problem with that proposal is that the maintenance of the HTML  
file and the maintenance of the server version is performed by  
different people, and that means it is impossible that they are  
consistent ;-)

What about adding a version parameter to each response from the  
endpoint?


On Nov 13, 2006, at 16:18, Recordon, David wrote:

> Solution 1 seems like the most simple.  "openid2" is a bit ugly, but
> does resolve the problem.  Otherwise we'd have to do something like
> Yadis discovery against the server endpoint which would make things  
> more
> complicated and meta.
>
> --David
>
> -----Original Message-----
> From: joshhoyt at gmail.com [mailto:joshhoyt at gmail.com] On Behalf Of Josh
> Hoyt
> Sent: Monday, November 13, 2006 4:15 PM
> To: Recordon, David
> Cc: specs at openid.net
> Subject: Re: Went Through it With Brad
>
> On 11/8/06, Recordon, David <drecordon at verisign.com> wrote:
>> 2) 7.3.3 basically deprecates HTML-based discovery, saying that it is
>> a way to know that the IdP is using Auth 1.1.  While I know we  
>> believe
>
>> Yadis will be used in most applications, I hypothesize that the
>> simplicity of HTML-based discovery will have it continue to prevail.
>> I thus would propose we remove the sentence saying that this is a way
>> to know that an IdP is running version 1.1.
>
> Yeah, it does. The justification for this is that there is no way to
> specify a version for the server, so we have to assume something, and
> since HTML discovery already used in 1.1, that's the only reasonable
> assumption to make. I see two ways out of this:
>
> 1. Add another "rel" value to the HTML discovery for OpenID 2:
>   <link rel="openid.server openid2.server" href="...">
>
> 2. Add some way of doing discovery on the endpoint URL for determining
> the version, so it doesn't have to be part of the user's XRDS or HTML
> document
>
> Either one of these would let us keep the nice, simple HTML discovery
> mechanism for 2.0.
>
> Thoughts or ideas?
>
> Josh
>
> _______________________________________________
> specs mailing list
> specs at openid.net
> http://openid.net/mailman/listinfo/specs