[foaf-dev] Re: Social Networks that use FOAF

[Peter Williams]

Having played with the work of others who _think_ "in the openid paradigm",=
 Im not sure I agree with your egneral openid sentiment.

- users and designers certainly do not express their worldview in terms of =
graphs. So RDF religion is not "embedded" at the doctrinal level.

- the transitive nature of the core FOAF relation (and the own functional m=
appins to owl:id properties) is also not addressed directly - thats for FOA=
F to do for HTTP URIs. For XRI-form URIs, a competing contact management se=
rvice (with n social graphs) is apparantly emerging. Its hard to say that o=
penid planned this - it simply comes with the adoption of the XRI name-form=
 as an openid.

- openid folks seem to have agreed that its not for openid to standardize a=
ny kind of trust model for parties, for attrbute flow, or security. Its for=
 the likes of the FOAF community to agument openid with policy controls, us=
ing what ever policy calculus (and a practical albegra, please!) that folks=
 come up. In the security literature, there are hundereds of proposals for =
infoflow logics, developed over the last 20 years. Some of them (sutherland=
, denning) are more respected than others: all are pretty academic - or nev=
er left military/natsec applications. (Of course, quad stores are included =
in this set.)



From: Phillip Rhodes
Sent: Mon 3/24/2008 9:35 PM
To: foaf-dev at lists.foaf-project.org
Subject: Re: [foaf-dev] Re: Social Networks that use FOAF


Story Henry wrote:
> To me it is quite clear why foaf is extreemly useful. I believe the =

> privacy issues can be dealt with quite simply with openid.

OpenID by itself isn't enough, there has to be some way to specify
policy around the info expressed in the FOAF.  Eg, maybe Susy can see
subset X of my "social graph" while Bob can see subset Y and Dave
can see anything.

FWIW, working up a way to express and control that kind of "policy =

stuff" is an explicit goal for OpenQabal.  I'm not far enough long
yet to have done any actual coding around that, but some ideas
are percolating in my mind.  Would be interested to see what
you lot think about how to deal with this point...


TTYL,

-- =

Phillip Rhodes
Chief Architect - OpenQabal
https://openqabal.dev.java.net
LinkedIn: http://www.linkedin.com/in/philliprhodes
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.usefulinc.com/pipermail/foaf-dev/attachments/20080325/3c2=
27dc5/attachment.htm