[OpenID] host-meta and "acct:"

[John Panzer]

Use case:

I have 1M user accounts, each identified with a URI.

I follow each URI to its associated XRD when I want to find out metadata
about that resource.

If I can do 2 things, I can avoid needing to publish (and digitally sign!)
1M XRD documents, by creating a "template XRD":
(1) My URIs inside the XRD can contain {uri}
(2) {uri} means, in this context, "The Subject of this XRD"
(3) My XRD has no Subject element, so I use as its Subject the URI (user
account) I started discovery on.

I can then sign exactly one document (the "template" XRD), and point to it
from all of my 1M resources. I mark it as cacheable for 1 week and it is
used for billions of metadata lookups without any HTTP overhead.  Life is
good.

(This is a profile of using XRD + templates of course, but a reasonable one
suggested to me by Breno.)

On Thu, Nov 5, 2009 at 3:30 PM, Manger, James H <
James.H.Manger at team.telstra.com> wrote:

>  Hi Santosh,
>
>
>
> > You are assuming the meta-data "is not a resource".
>
>
>
> No I am not.
>
>
>
> > I think meta-data is also a "Resource".
>
>
>
> I agree, the jim.footy.xrd  file has its own URI so it is a resource — but
> that is irrelevant to <Subject>.
>
>
>
> The XRD <Subject> element “provides the identifier of the resource *described
> by* this XRD” [XRD v1.0 WD9, §2.1.1.<http://www.oasis-open.org/committees/download.php/34724/xrd-1.0-wd09.html#element.xrd>
> ].
>
> jim.footy.xrd does not describe itself, it describes hundreds of other
> *.html pages I wrote about football. It makes no sense for <Subject> to hold
> the URI of the XRD file itself in this situation.
>
>
>
>
>
> If you have an OpenID identifier that isn’t used for anything else (eg its
> not your blog or home page), then (in some potential future OpenID v3) a GET
> on that OpenID identifier could return a XRD with its own URI in the
> <Subject>. It would be metadata about itself. It could point to an OP
> directly, eliminating the extra round-trip of going from a resource to its
> metadata since they are one and the same in this case. [Mind you, I’m not
> sure you gain much over serving an HTML page with OpenID v1 or v2 <link>s to
> the OP.]
>
>
>
>
>
> James Manger
>
> _______________________________________________
> general mailing list
> general at lists.openid.net
> http://lists.openid.net/mailman/listinfo/openid-general
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openid.net/pipermail/openid-general/attachments/20091105/666a9ada/attachment.htm>