Jens,
Well, thats something a lot of people are trying to figure out, in
broader terms; what areas of "identity" is seperated into distinct
services? How are things kept seperate, yet aggregated, and then how
are they secured and protected so that we can show only certain
parties certain things? The FOAF people have one answer, the SIOC have
a slightly different answer, etc. I had been working early on with the
dp.org group (when it was sub 30 people, before scoble-gate) on some
reference stuff, and then when that got too big, we spun it off into
"WRFS".
http://groups.google.com/group/wrfs
Its essentially a "discovery / aggregation / security " protocol using
openID, AX, OAuth+Discovery to allow a decentralized mechanic for
using your data in whatever 3rd party app you like, regardless of
location or API type.
Initial sketch document:
http://cowbell.floe.tv/WRFS_11_20_2007.html
We're sorta in stealth mode, getting a functional prototype up based
on a working document:
http://wrfs.googlecode.com/svn/trunk/Docs/Draft1/WRFS_Spec_MDraft_v0_1.xml
(run that through
http://xml.resource.org/ to get html)
and evolving it as use cases illustrate need.
If you are interested in working on the prototype, and then eventually
the spec, join up with our group.
Josh Patterson
> It makes sense forhttp://
axschema.org/contact/IM/*, as their use
> requires different protocols and only stores usernames, and it could
> make sense to use an attribute for e.g. flickr usernames.
>
> But I don't see an advantage in providing attributes to store site
> specific urls (who has to check the consitency anyways), these urls
> should be stored in more generic attributes, likehttp://
axschema.org/contact/web/photo.