Server-to-server channel
Chris Drake
christopher at pobox.com
Mon Apr 2 18:50:46 UTC 2007
Hi,
I've been away a while - is there any server-to-server mechanism built
into OpenID yet?
I've noticed folks wanting to build a "log off" facility for OpenID,
which essentially requires the users server to inform whatever places
the user has been recently to "drop" any session info.
I've also noticed a lot of discussion about attributes, which begs the
question about how to handle things that change (eg: If I've given out
my email address to a dozen web sites, and then I change it, how does
my OpenID server propagate that change to all those sites?)
"User Centric" implies that sites don't store anything about me, and
that whenever they need to know stuff (eg: my email), they instead ask
my OpenID server, which returns them the answer (unless I've since
revoked permission or whatever). Again - server-to-server (although
this time in the reverse direction) applies here.
Many months ago, I proposed the idea of "Single Sign On" - which is
defined as letting a user access one or more web sites in some short
period of time, without that user having to type anything in (not
their password, not their username/email, not even their giant long
identity URL), and while letting that user click just once (or
preferably zero times - that is - they're auto-magically "single
signed on" as soon as they re-visit the next different
OpenID-compatible web site for the day) to "get in". server-to-server
is again required to accomplish this user-friendly functionality.
Since I've been offline, I'm confident that there have been more use
cases for server-to-server proposed by other people as well.
Is there any way for providers and consumers to identify one-anothers
endpoints yet (in the "absence" of the user's browser as a transport
mechanism), and for attributes etc to be exchanged between?
Kind Regards,
Chris Drake
More information about the specs
mailing list