- From: Rigo Wenning <rigo@w3.org>
- Date: Mon, 7 Mar 2005 15:21:15 +0100
- To: "Giles Hogben" <giles.hogben@jrc.it>
- Cc: public-p3p-spec@w3.org
Received on Monday, 7 March 2005 14:21:24 UTC
Am Wednesday 23 February 2005 12:15 verlautbarte Giles Hogben : > What is the difference between. > "assume that all data is collected" > and > "any data may be collected" The issue with this is, that from a risk perspective (client-side), they are the same for P3P. Unless you have a wallet within your P3P client, you may not know how and what they collect direct from you or from protocol information etc. Additionally, it gives services to get away with a huge list of collections but hiding behind the "may". Saying, we collect race, political opinion, etc is different from saying: we may do that in some exceptional circumstances and we may (or may not) use it. In this case, you can't know what is collected and used as the "may" opens the can and everything "may" be collected, but P3P has rather to look into what is "actually" collected. > > This is what I meant... > This is a modal logic problem (the logic of possibility) > See http://plato.stanford.edu/entries/logic-modal/#1 Problem here is that marketing isn't always following those thoughts ;) Rigo
Received on Monday, 7 March 2005 14:21:24 UTC