During the last decades, interdisciplinary research has gained significant popularity. While scholars typically work in their own self-contained and isolated domain within their community of experts, we refer to interdisciplinary research as that which brings together approaches of at least two different disciplines.

In this work, we report on our interdisciplinary research on the protection of location data. We tackle the problem from an engineering, legal, and ethical disciplinary perspective. While there are several reasons why interdisciplinary research can be fruitful,[1] we think interdisciplinary research is particularly useful for matters regarding data protection. In order to understand how data is created, transmitted and processed, one needs an understanding of technical systems, i.e. from the perspective of engineering. Yet, data processing is not merely a matter of technical possibilities, but also one of legal regulation. Hence one needs knowledge from the legal domain. Finally, we use Helen Nissenbaum’s Contextual Integrity (CI) heuristic,[2] based on an ethical approach, as a middle ground between legal and technical assessments of privacy violations.

People use their smart devices, with their positioning capabilities, to engage in a wide variety of location-based services (LBSs). These services have in common that users must share their current whereabouts with a service provider to, for example, find nearby points of interest, share location data with friends, or get directions. It is well known that the ensuing mass dissemination of location data generates significant privacy concerns because location data reveals information about users that is potentially sensitive, difficult to anonymise,[3] and entities with access to accurate location data are able to make inferences about, for example, home/work address, income level, religious beliefs, sexual preferences or health issues.[4] To make things worse, behind the scenes, users share their location data with many more entities than they may be aware of and their location data may be used for purposes that they would never anticipate. This is mainly due to the current business model of many LBSs. In the case of free services, service providers finance their service by either adding third party advertisers to their applications or by selling user data to data brokers.[5] Note that LBSs thus collect location data that is not necessary to deliver their service.[6]

This work is, to the best of our knowledge, the first that tackles the protection of location data from an engineering, ethical and legal perspective. From an interdisciplinary perspective, our article has four main contributions: first, the technical detail from an engineering perspective provides a substantive added value in connection with the ethical as well as the legal discipline. Second, our article serves as a reference for scholars of the involved disciplines to learn how the issue is addressed in the other two disciplines. Third, we identify a special relation between the ethical and the legal discipline, i.e. the connection between the concept of contextual integrity and purpose limitation. Fourth, our article serves as a case study on how to do interdisciplinary investigations of a data privacy matter. Additional to these interdisciplinary contributions, our work also provides valuable contributions to the CI heuristic and to the connection between the CI heuristic and data protection law. We show how the CI heuristic can be applied in a way that sensitises readers (and users) to what is at stake, and clarifies what the heuristic adds to the commonly stated opinion that location data can be sensitive data. Finally, our article discusses the legal concept of purpose limitation with respect to location data and argues its added value compared to contextual integrity.

Many of the terms used in this work have a precise meaning within one discipline, while evoking less precise connotations within the “other” discipline. For instance, in legal terms, sensitive data refers to a specific category of data, summed up in art. 8 of the Data Protection Directive (DPD) and in art. 9 of the General Data Protection Regulation (GDPR) that will replace the DPR from May 2018.[7] This concerns personal data revealing, among other things, ethnic or racial origin, or political beliefs. Being qualified as sensitive data has legal effect, since the processing of such data is by default prohibited. Location data is not sensitive data in this sense, though individuals may perceive their location to reveal sensitive information in a more general sense, and when correlated with other data location data may indeed result in data that is “sensitive” in the sense of EU data protection law. Since this article is co-authored by computer engineers and lawyers, we refer to sensitive data in the general sense of the term and will specify when we use the term in the legal sense. Other examples of potential misunderstandings may arise, for instance, when engineers speak of “users”, “clients”, and “service providers”, whereas lawyers speak of “data subjects” and “data controllers. This is important because legal terms have legal effect and must therefore be used with precision. By specifying the legal meaning whenever relevant, we aim to contribute to the necessary dialogue between both disciplines on the challenges and solutions regarding the proliferation of location data.