Shibboleth Developers

[Scott Cantor <>]

> > I'm open to moving this to a separate document, I guess. What do 
> > others think?
> 
> I don't think this is necessary, if we just modify the 
> existing material to emphasize we're illustrating desirable AA
> capabilities, not mandating an implementation.

Unfortunately, my revisions of section 5.6 are creating a much more
specific explanation of *this* ARP processing model, precisely because
we've had so much trouble understanding it and because the document as
it stands was intended to call out that model.

I'm not changing the authors' intent by making it more explicit. I've
confirmed that at length in talking with Marlena. Her original text in
fact, is closer to my new text than to rev. 4.

This would be a change to a "non-normative example" style. I definitely
don't want to lose some of what's there either, but the document is
filled with (SHAR, URL, Release) triplets in both sections 3 and 5.
That's a specific kind of ARP, even without these processing rules. If
it's not the only kind we want, then we shouldn't go on at such length
with those examples.

I don't fundamentally see a problem with saying, like we do with AAP's,
"these things are internal to the AA, the input to the policy evaluation
process consists of SHAR, URL, and handle, this is what one might look
like and here are some issues surrounding them". The rest is left to the
imagination or other documents. But it is a change.

-- Scott

------------------------------------------------------mace-shib-design-+
For list utilities, archives, subscribe, unsubscribe, etc. please visit the
ListProc web interface at 

    http://archives.internet2.edu/

------------------------------------------------------mace-shib-design--