9 Squares: Framing Data Privacy Issues

In order to frame discussions on data privacy in varied contexts, this paper introduces a categorisation of personal data along two dimensions. Each of the nine resulting categories offers a significantly different flavour of issues in data privacy. Some issues can also be perceived as a tension along a boundary between different categories.

The first dimension is data ownership: who holds or publishes the data. The three possibilities are “me”, i.e. the data subject; “us”, where the data subject is part of a community; and “them”, where the data subject is indeed a subject only. The middle category contains social networks as the most interesting instance. The amount of control for the data subject moves from complete control in the “me” category to very little at all in the “them” square – but the other dimension also plays a role in that.

The second dimension has three possibilities, too, focusing on the type of personal data recorded: “attributes” are what would traditionally be found in databases, and what one might think of first for “data protection”. The second type of data is “stories”, which is personal data (explicitly) produced by the data subjects, such as emails, pictures, and social network posts. The final type is “behaviours”, which is (implicitly) generated personal data, such as locations and browsing histories. The data subject has very little control over this data, even in the “us” category. This lack of control, which is closely related to the business models of the “us” category, is likely the major data privacy problem of our time.