This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5).
Online social networks (OSNs) have become central to the lives of millions of people worldwide. Unfortunately, existing OSNs cede responsibility for user data to a single administrative entity, and are inherently prone to violations of usersʼ privacy. This sensitive user data creates an attractive target for hackers and can be abused by internal administrators. This work argues that more decentralized alternatives to the dominant OSN architecture can provide a better balance between features and privacy. By endowing OSN participants with full ownership and control of their personal data, including control over which machines are allowed to store information and who is allowed to access it, decentralized OSN architectures have the potential to reduce the risk of a large-scale privacy breach while providing OSN features, efficiency, and availability that are competitive with more centralized schemes.
The work will provide insights into the fundamental tension between privacy and features in OSN-service architectures. Implementation and evaluation of methods based on these insights will lead to greater understanding of the features, efficiency, and availability that decentralized OSN architectures can support. The work will enable scientific inquiry into a wide range of social phenomena through the development of privacy-preserving methods and infrastructure for collecting location data. The work will also strengthen ties between computer science and on-campus initiatives for integrating undergraduate education and research through the Duke SmartHome.
