Designing Privacy-Enhanced Interfaces on Digital Tabletops for Public Settings

Abstract

Protection of personal information has become a critical issue in the digital world. Many companies and service provider websites have adopted privacy policies and practices to protect users’ personal information to some extent. In addition, various governments are adopting privacy protection legislation. System developers, service providers, and interface designers play an important role in determining how to make systems fulfill legal requirements and satisfy users. The human factor requirements for effective privacy interface design can be categorized into four groups: (1) comprehension, (2) consciousness, (3) control, and (4) consent (Patrick & Kenny, 2003). Moreover, the type of technology that people are engaged with has a crucial role in determining what type of practices should be adopted. As Weiser (1996) envisioned, we are now in an “ubiquitous computing” (Ubicomp) era in which technologies such as digital tabletops (what Weiser called LiveBoards) are emerging for use in public settings. The collaborative and open nature of this type of smart device introduces new privacy threats that have not yet been thoroughly investigated and as a result have not been addressed in companies’ and governmental privacy statements and legislation. In this thesis, I provide an analytical description of the privacy threats unique to tabletop display environments. I then present several design suggestions for a tabletop display interface that addresses and mitigates these threats, followed by a qualitative evaluation of these designs based on Patrick and Kenny’s (2003) model. Results show that most participants have often experienced being shoulder-surfed or had privacy issues when sharing information with someone in a collaborative environment. Therefore, they found most of the techniques designed in this thesis helpful in providing information privacy for them when they are engaged with online social activities on digital tabletops in public settings. Among all of the proposed tested designs, the first three have proven to be effective in providing the required privacy. However, designs 4 and 5 had some shortfalls that made them less helpful for participants. The main problem with these two designs was that participants had difficulty understanding what they had to do in order to complete the given tasks.