Enabling People-Centric Sensing by Overcoming the privacy BarriEr (SNSF-149591)
Ubiquitous data collection from our smartphones is becoming a mainstream. Some data (e.g., photos, videos, location, sentiments) we share explicitly with our social networks, or with owners of data collection campaigns serving research in, for example, human-computer interaction, sociology, smart cities and so on. Additionally, some data is collected automatically by applications running on our smartphones to offer us more tailored services, for example location-based search for nearby coffee places.
It is becoming increasingly difficult for us, the smartphone users at large, to control who collects which of our personal data, when they collect it, and for what purpose. Imagine that you set up your Google glasses to automatically share photos with your friends and some work colleagues. One day you are attending a family meeting, and many funny moments occur; yet you wish to keep these in the family circle (i.e., outside that context they can be embarrassing or show "your real self"). Therefore, you want to take photos with your glasses to remember that day but you wish to share them only with your family. In the worst case you forget your glass settings and all the photos are shared with others. In the best case, you need to remember to manually adjust the sharing options. This is obviously tedious.
This proposed PCS-OBEY project researches these types of challenges and defines a theoretical user model and prototyped framework to design semi-automated privacy settings mechanisms to help users like you to better manage how their private data is collected and with whom it is shared depending on their contextual situation. For the above example, the PCS-OBEY solution would share pictures only with your family members who are part of the context in which these photos are taken. Besides the users, developers of new mobile applications and services will benefit from PCS-OBEY because it will be easier for the users to use and trust their revolutionary ubiquitous features.