Mobile phones has become ubiquitous personal devices. In our daily lives we use them to communicate with people access digital worlds trough networks. We also use them as sensors and connectors to our nearby physical environment to use them for navigation, counting steps, or connecting to the TV or stereo. But so far we have not been able to sense people in proximity. Our phones could form the link between our physical and digital selves, also to other people. This would enable a new type of applications.
The Proximates project studies pervasive social context across heterogenous and dynamic social networks, including physical proximity networks.
More concretely we aim at building a social context engine, an Android component that provides authorized apps with information about the social context of the user, in a way that protects the user's privacy. By social context we mean wether a user is with friends, family or colleagues . Such a component can be used to build context aware applications. For example, users' call handling decisions depend heavily on their social context.
So far we have used the Proximates component to build some prototype applications:
Previous studies, e.g. The MIT Reality Mining project, have been limited in size and use surveys and controlled user groups, typically students. In Proximates we want to investigate if this kind of study can be scaled to larger user groups without using surveys, and instead use Facebook. The Memorit application has been developed in this project to facilitate this. We also have a control group of about 50 users that we use as reference group.
In the project, we also study privacy issues. We have applied Akerlof's lemons market game theoretic model to app stores which explains why users don't read Terms of Service.
This led us to see informed consent as a process rather than a transaction, and resulted in a design guideline for application developers to ensure that users stay informed and consenting: The Obvious Data Usage guideline.
Memorit is available for download from Google Play: https://play.google.com/store/apps/details?id=se.lth.cs.memorit
Please support our project by downloading and using Memorit.
If you are interested in Proximates you may also be interested in the Device Analyzer project at University of Cambridge: http://deviceanalyzer.cl.cam.ac.uk/