Fwd: Yadis

[Ben Nolan]

Forwarding Brad's reply to my email.

>> I'll post to the mailing list when I've got a server working.

I've got the yadis client going at foopad.com <http://foopad.com> - 
hopefully have it public tonight. There's some cool stuff you can do with 
the users foaf file too.

>> The idea looks excellent, although I'd love to be able to use the system 
to
>> give a site permission to write to my metaweblog API. I've got some ideas
>> about how this could be added - but it's slightly complex.
>
>I've been idly thinking about the same sorts of things... grant
>permissions along with the assertions or something.

One way would be to create a new param on the call to the ID server:

is_foaf=http://www.livejournal.com/..../foaf
has_permission_to=//xpath/to/xml_endpoint/in/foaf/file

Then when the ID server has confirmed the user wants to let
foopad.com<http://foopad.com>access their endpoint - the ID server
sends a message to the xml endpoint
(assuming you control it or can message it) saying 'let
http://foopad.com/write to the blog'. Then when
foopad.com <http://foopad.com> opens a metaweblog session, it sends the foaf 
url and a signature signed by the foopad.com <http://foopad.com> DSA key.

>> What about sixapart and typekey - are they aware that you're making 
typekey
>> redundant?
>
>I'm pushing for both TypeKey to speak Yadis (trivial, trival addition---
>they already do DSA signatures and FOAF files) as well as Movable Type
>being a Yadis consumer and server. (so MT users can be their own identity
>authority).

Sounds great.

Regards,
Ben
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.danga.com/pipermail/yadis/attachments/20050517/b3fa8771/attachment.htm