Being Forgotten on the Internet: How Temporal Contextual Integrity can Protect Online Reputation

On the web, personal information is retained forever. Facebook stores every "like", and Google every search query that was ever made. At the same time, people's lives are constantly progressing and advancing, resulting in a tension between how people are and what the retained data tells about them. Recently, this tension surfaced in the discussion on a "forgetting Internet" reverberated in the cases of Stacey Snyder v. Millersville Univ. (07-1660 E.D.P. 2008) and Google v. Mario Costeja González (ECJ C-131/12). Snyder was refused getting a job as a teacher because the web remembered social media content from college showing her drunk. And Google had to remove twelve year-old web search entries attesting Mr. González a bad creditworthiness. People are deprived of getting a second chance and to start a new life because the aspect of time lacks prominence in how people are presented online. The web is short of temporal context and sheds a distorted light on people's biographies. Aiming to restore a truthful picture of people's biographies online, this thesis adapts Nissenbaum's theory of privacy as contextual integrity and, based on Heidegger's and Ricoeur's philosophy of time, Walzer's spheres of justice and Solove's visionary pragmatism for privacy, proposes protecting the temporal contextual integrity of personal information. Temporal contextual integrity of a person's online profile means that it is not distorted by outdated personal information. Combining experimental, survey, focus group, expert, and usability methods, design-oriented and behavioral research on preserving temporal contextual integrity is conducted. The thesis develops user interface components making the passage of time within people's online representations prominent. Whether time-sensitive interface components displaying people's biographies increase their chances of getting hired in online labor markets despite an obsolete reputation is then investigated in an experimental study.