- From: Rigo Wenning <rigo@w3.org>
- Date: Fri, 22 Jun 2001 20:37:44 +0200
- To: www-p3p-dev@w3.org
- Cc: cbf@profileup.com
Hi Christophe, We come from a user-perspective in this question. If you would cite the whole sentence, it would be much clearer. This addresses mostly the client-side implementation. And there it says: If a user agent discovers more than one non-expired P3P policy for a given URI..[1]. I don't see the user-agent now checking all the headers of a site for any potential URI on that site. This can be infinite. So the emphasis is on "discover". And this is under an "if". So there is no rule, that obliges a P3P client to scan all the potential headers of a web-site. From a server-side perspective, it is no problem either, as you have control over your configuration and you SHOULD avoid to have more than one PRF covering the same URI. How to configer servers can be found in the implementation Guide[2] Only to prevent a theoretical aproach, there can not be two head-answers for one http-request. So I think, that finally, the risk is on the server side. In the case of declaration of multiple policies, the spec make's the assumption, that the declaring party follows all the rules, it declares. This means, that if there are two conflicting policies, you merge them and can only collect/use data for purposes, which are allowed under both policies. I hope that helps Rigo Wenning W3C/INRIA Policy Analyst Privacy Activity Lead mail:rigo@w3.org 2004, Routes des Lucioles +33 (0)6 73 84 87 31 F-06902 Sophia Antipolis http://www.w3.org/ On 01-06-21 12:02, Christophe Brun-Franc wrote: > Hi > > > Doest this sentence : > > "because P3P headers for two pages on the site reference > different policy reference files that declare different > policies for the same URI) > " > > means that we have to analyse all HTTP headers of the entire web site to be > sure that we get all the different policies for an uri ? > - Of course, except if there is a reference file in a well-known location - > > It's seems where difficult to do that ... > 1. http://lists.w3.org/Archives/Public/www-p3p-dev/2001Apr/0001.html 2. http://www.w3.org/TR/p3pdeployment
Received on Friday, 22 June 2001 14:37:51 UTC