Moving AX Forward (WAS RE: SREG namespace URI rollback)

[Recordon, David]

> Actually it is describing a document format, and it could easily be
used
> by other groups as evidenced by references from people in the ID
Schemas
> group.
I agree that it could be, but is anyone?  I love shooting beyond the 80%
to get the remaining 20%, but if that is just a pipe dream then I have a
hard time seeing why the documents need to be separate and thus more
complex.  If however this format was defined within the ID Schemas
project, then that would be an easy argument as to why they should be
separate.

> We defined a set of attributes 6 months ago under schema.openid.net.
So Dick, this is part of my problem with AX.  Sxip has defined a set of
attributes and never gained consensus on this list that that is the
right thing to do.

See my other message a few minutes ago as to the rest of my thoughts.

--David

-----Original Message-----
From: Dick Hardt [mailto:dick at sxip.com] 
Sent: Thursday, April 05, 2007 8:27 AM
To: Recordon, David
Cc: Johnny Bufu; OpenID specs list
Subject: Re: Moving AX Forward (WAS RE: SREG namespace URI rollback)


On 4-Apr-07, at 1:16 PM, Recordon, David wrote:

> Johnny,
> I see a lot of, at least my initial confusion, coming from there being

> multiple documents.  This is why I urge merging the transport and 
> metadata since the reality is they currently are only being used with 
> each other.  As the metadata document doesn't actually define a new 
> format, rather references existing formats, I am unsure why it cannot 
> just be a section in the transport document.  It is understood that 
> you must use the metadata format for the schema URLs in the transport,

> so the two documents really are coupled to begin with.

Actually it is describing a document format, and it could easily be used
by other groups as evidenced by references from people in the ID Schemas
group.

>
> I agree that you need to bootstrap a set of attributes for people 
> using AX.  As I have done so in the past, I'd encourage this work 
> happen within the ID Schemas project (http://idschemas.idcommons.net/)

> versus defining First Name yet again for openid.net.  I have no 
> problem with the spec listing a set of schema URLs, I just strongly 
> feel that anything non-OpenID specific should be hosted and defined 
> elsewhere since so many people have already done it.  I do understand 
> the need for the schema URL hosting the metadata document, which is 
> why I am advocating this work be done as part of the ID Schemas 
> project to provide this flexibility.

see my response to Drummond ...

We defined a set of attributes 6 months ago under schema.openid.net.

I think we have let other groups have time to do something, I'd like to
get on with building and deploying stuff.

-- Dick