A Formal Privacy Management Framework

Abstract

Privacy is a complex issue which cannot be handled by exclusively technical means. The work described in this paper results from a multidisciplinary project involving lawyers and computer scientists with the double goal to (1) reconsider the fundamental values motivating privacy protection and (2) study the conditions for a better protection of these values by a combination of legal and technical means. One of these conditions is to provide to the individuals effective ways to convey their consent to the disclosure of their personal data. This paper focuses on the formal framework proposed in the project to deliver this consent through software agents. Context and MotivationsIn the same way as the growing use of photography at the end of the 19th century prompted Warren and Brandeis seminal paper [31], the changes imposed nowadays by information and communication technologies require a deep reflection on the fundamental values underlying privacy and the best way to achieve their protection [15,27]. Furthermore a multidisciplinary approach is necessary to tackle this challenge because privacy can neither be apprehended nor guaranteed by exclusively legal or technical means. As a step in this direction, the collaborative project PRIAM 1 gathers lawyers and computer scientists with the goal of putting forward a common view of privacy for pervasive computing and effective (legal and technical) instruments to protect it.

Accountability for data protection

“…Several articles already mention the limits of the pure security approach (for instance [19]) and suggest to rely on the concept of accountability. Thus, there is a recent interest and active researches for accountability which overlap several domains like security [19,20,21], language representation [22,13], auditing systems [8,23], evidence collection [15,14] and so on. However, only few of them consider an interdisciplinary view of accountability taking into account legal and business aspects.…”

Privacy by design

“…• Language based approaches: a number of languages and logics have been proposed to express privacy policies [1, 3,4,5,11,12,13,21,23,30,24,35,38,53]. These languages may target citizens, businesses or organizations; they can be used to express individual privacy policies, corporate rules or legal rules.…”

“…[1, 3,4,5,30,21,35,38,53]), they can be used to verify consistency properties or to check if a system complies with a privacy policy. These verifications can be performed either a priori, through static verification techniques, on the fly, using monitoring, or a posteriori in the context of audits or accountability procedures.…”

Privacy Verification Using Ontologies

“…Verification of policies uses a static control/data flow analysis. Métayer proposes a formal framework to deliver the individuals consent regarding the processing of its personal information through software agents [17].…”