Re: expiry doubt

The spec does seem to be ambiguous on this point. The
P3P spec group will look into a clarification for P3P1.1.
In the mean time, my advice is to be conservative and assume
a 24 hour expiry when in doubt.

Lorrie


----- Original Message -----
From: "Elena" <elena.dasseni@txt.it>
To: <www-p3p-dev@w3.org>
Sent: Thursday, July 11, 2002 6:13 AM
Subject: expiry doubt


Hi,
    I've got a question on the expiration date of the policies. What I
evinced from P3P specs is what follows:

1) Policy Reference File : if the POLICY-REFERENCES node has the EXPRY node
as a child, then - for all the policy references that follow - the
expiration date is computed using the info carried in the EXPIRY node.
Otherwise, such policy references have a 24-hours validity.
2) Policy: if the POLICIES node (that MUST appear in any xml file
representing a policy) has the EXPRY node as a child, then the policy
expiration date is computed using the info carried in the EXPIRY node.
Otherwise, the policy has a 24-hours validity.

So, in both cases, we have 2 ways to compute the expiration of a
policy/policy ref file.
I've encountered a Policy Ref File with the following structure, that made
me think about a third way:

<POLICY-REFERENCES>
<EXPIRY max-age="86400 " />
<POLICY-REF about="#XYZ">
<INCLUDE>/*</INCLUDE>
</POLICY-REF>
</POLICY-REFERENCES>
<POLICIES xmlns="http://www.w3.org/2002/01/P3Pv1">
<POLICY name="XYZ">
.....
....
</POLICY>
</POLICIES>

The POLICIES node doesn't have an EXPIRY node, so - according to P3P
Recommendation - all the POLICY it embeds should be treated as having a
24-hours validity.
However, I was wondering if - since the POLICIES node is inside the Policy
Reference File - we should use the EXPIRY node inside the POLICY-REFERENCES
to compute the POLICY validity.
Maybe this sounds crazy, and it doesn't make great sense to me too, but I
would like to have this point clear.

Thanks a lot

Received on Friday, 12 July 2002 10:15:46 UTC