X-Original-To: pgsql-admin-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id D17898B9CA0; Tue, 22 Feb 2005 20:51:13 +0000 (GMT) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) with ESMTP id 05891-08; Tue, 22 Feb 2005 20:51:11 +0000 (GMT) Received: from koolancexeon.g2switchworks.com (mail.g2switchworks.com [63.87.162.25]) by svr1.postgresql.org (Postfix) with ESMTP id 519FF8B9D37; Tue, 22 Feb 2005 20:51:05 +0000 (GMT) Received: mail.g2switchworks.com 10.10.1.8 from 10.10.1.37 10.10.1.37 via HTTP with MS-WebStorage 6.5.6944 Received: from state.g2switchworks.com by mail.g2switchworks.com; 22 Feb 2005 14:51:00 -0600 Subject: Re: [GENERAL] How to Prevent Certain Kinds of Joins? From: Scott Marlowe To: Heather Johnson Cc: Michael Fuhr , pgsql-admin@postgresql.org, pgsql-general@postgresql.org In-Reply-To: <421B9674.4060807@nypost.com> References: <421B8D2D.6050800@nypost.com> <20050222200915.GA71591@winnie.fuhr.org> <421B9674.4060807@nypost.com> Content-Type: text/plain Content-Transfer-Encoding: 7bit Message-Id: <1109105459.30529.119.camel@state.g2switchworks.com> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.6 (1.4.6-2) Date: Tue, 22 Feb 2005 14:51:00 -0600 X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=0.001 tagged_above=0 required=5 tests=AWL X-Spam-Level: X-Archive-Number: 200502/235 X-Sequence-Number: 16403 On Tue, 2005-02-22 at 14:30, Heather Johnson wrote: > One of the tables contains personally identifiable information (PII) > (e.g., email, first and last name, etc.). The other contains "click > stream" data (information about where on our website users have gone). > In order to be sensitive to users privacy, we don't want to ever (even > accidentally) run queries that would combine PII and click stream data. > So we're looking for ways to put "walls" up against combining the data. > We understand that anyone with ample access to the database can > deliberately combine the data if they chose to do so in blatent > violation of our policies. But we want to set something up that would > put an obstacle or two in the path of anyone who might accidentally run > such a query. > > The ideas about setting up views might work. I guess we'd just have to > change the permissions of users who have access to the db so that they > can only use the views and not query the tables directly. Why not change the keys that currently connect them to something different (i.e. random noise) and make a NEW table that could join them with those random keys that is restriced access wise to only the chosen few. Or do you need to actually ever re-reference the two datasets? If not, then just lose the connecting data when you insert the rows.