1) (Based on the incredibly small amount i know about Open ID) it’s not a
new idea… the concept of “single sign-on” has existed since the birth of
the internet - (but you knew that already — let us not forget MS Passport
and the tragic, rediculous, irritating failure that was.) This bandwagon
kicks up every few years - ends up too hard for people to deal with because of standards and proprietary issues - and then retreats again back into the murky land of excellent ideas that may never come to fruition. (Even I helped to write a business plan that was aiming for this same target 8-9 years ago.)

The problem with single sign-on - or single source account management - is
just that… it’s a singular place to go. It’s essentially a single point of
failure. Something goes wrong (either user failure or some enterprising
hacker) and everything goes to hell. So far there’s been no single
destination, company, or entity that people have trusted to be the
gatekeeper to their total data store. Not saying that there couldn’t be
(google seems to have been gaining in trust over the past couple years), but
it’s a big barrier to entry for people. Very few people like having all
their eggs in one basket.

2) Your last point about the social networking perspective is an intriguing
one - and the specific observation about maintaining different types of
profiles on different portals is absolutely true. My LinkedIn profile looks
way different than my facebook profile - and for good reason.

It seems to me that ultimately you’ll want some level of “profile” management. You’ll want some way to manage a single pile of stuff about yourself then dole that information out to appropriate destinations. As you said — where do you go for that? who maintains that? Obviously this is where Ringside could come
in to play… However, what in this model is any different than what I do today? I
have a pile of info about myself (that presumably i keep on my local
machine) and i choose what bits i want to share and where i want to put
them. Seems to me that implementing profile management is a no brainer - but
how do you get people to want to use it? On the surface there’s not a whole
lot of value add — and in fact there only seems to be added risk? (again
single point of failure, privacy and trust issues, etc. etc. etc.)

Inspired by the the South African XMPP Federation OpenID Server, I wrote a case scenario of user-profile export to an OpenID/XMPP server. And this could be combined with OpenID Attribute Exchange then.

Going further, any webservice supporting XMPP could import user-profiles directly.

An OpenID/JabberID convergence, with an OpenID-to-JabberID delegation mechanism, would be very nice, though.