The Potential for User-Tailored Privacy on Facebook

Research shows that Facebook users differ extensively in their use of various privacy features, and that they generally find it difficult to translate their desired privacy preferences into concrete interface actions. Our work explores the use of User-Tailored Privacy (UTP) to adapt Facebook's privacy features to the user's personal preferences. We developed adaptive versions of 19 Facebook privacy features, and for each feature we test three adaptation methods (Automation, Highlight and Suggestion) that can be used to implement the adaptive behavior. In a "think-aloud" semistructured interview study (N=18), we show participants paper prototypes of our adaptive privacy features and ask participants to judge the presented adaptive capabilities and the three adaptation methods that implement them. Our findings provide insights into the viability of User-Tailored Privacy. Specifically, we find that the optimal adaptation method depends on the users' familiarity with the privacy feature and how they use them, and their judgment of the awkwardness and irreversibility of the implemented privacy functionality. We conclude with design recommendations for the implementation of User-Tailored Privacy on Facebook and other social network platforms.