> I poked around for SAML 1.1 standards docs, and I didn't find anything that
> differentiated between "Person Attributes" and "Authentication Attributes"
> like the CAS model does.

That was my only concern. We've been loose with SAML to date, and I'd
be concerned to further abuse the standard. If an attribute statement
doesn't strictly define attributes about the principal described by
the NameIdentifier in the Subject, then I have no concerns with the
approach. Indeed it would likely simplify a number of use cases
(including ours).

> Is this violating any design principles by
> attaching credential-derived attributes to the principal?

Not that I'm aware of.

M

-- 
You are currently subscribed to [email protected] as: 
[email protected]
To unsubscribe, change settings or access archives, see 
http://www.ja-sig.org/wiki/display/JSG/cas-dev

Reply via email to