RE: 2.0 draft 13 comments

[Tim Moses <tim.moses@entrust.com>]

Seth - The reason that the core spec. commonly qualifies <AttributeValue>
with the "xacml:" namespace is that <AttributeValue> is defined separately
and differently in both the xacml and xacml-context namespaces.  So, I
suggest rejecting your comment on Section 7.3.  Let me know if you disagree.
All the best.  Tim.

-----Original Message-----
From: Seth Proctor [mailto:Seth.Proctor@Sun.COM] 
Sent: Thursday, August 26, 2004 6:40 PM
To: tim.moses@entrust.com; xacml@lists.oasis-open.org
Subject: 2.0 draft 13 comments



Attached are my comments on draft 13 of the core 2.0 specification.

NOTE WELL: I haven't seen comments from anyone to suggest that they've read
the recent drafts cover-to-cover looking for issues, so I took the last two
days to do this. Since we're in a time crunch, however, I didn't stop to
scrutinize every last detail, so I make no guarentee that I caught all the
issues. Originally I was looking only at the issues I owned, so there is
some emphasis on that text. I don't believe I am suggesting any changes in
functionality or meaning (I tried to take all those issues to the list in
separate emails).

...Also, a seriously big thank you to Polar for helping me hash out some of
my questions at the last minute! In fact I've asked him to look at the
any-of-all and all-of-any functions too...

The one issue I don't think we have closure on is the mandate that combining
algorithms must return NotApplicable if their elements are all
NotApplicable. Once we clarify this, then I think all my comments will be
in.


seth