Dec. 15, 2010:

Confidant software is now released. Download and protect your online privacy!

Dec. 15, 2010:

Confidant website is up. Feedback and comments are welcome!


Online social networks (OSNs) are immensely popular, but participants are increasingly uneasy with their centralized control of user data. Decentralized OSNs offer the potential to resolve this tension while also enhancing the features and scalability offered by existing, centralized services.

Confidant is a decentralized OSN designed to enable scalable data-processing of OSN data without conpromising users' privacy. Unlike previous decentralized OSNs, which assume that storage servers are untrusted, Confidant replicates a user's data on servers controlled by her friends. Because data is hosted on trusted storage servers, Confidant can support a scalable data-processing framework for OSN data. The key challenge in realizing this vision is managing access-control policies under weakly-consistent replication. Confidant addresses this challenge by eliminating write conflicts through a lightweight cloud state manager and through a simple mechanism for updating the bindings between access policies and replicated data.


AT&T Labs
  • Ramón Cáceres
  • Alexander Varshavsky

Acknowledgements and Sponsoring

This work is funded with NSF Award CNS-0916649, "NetSE: Small: Privacy-preserving Architectures for Social Networking Services," as well as support from AT&T Labs and Amazon.

Duke Computer Science Logo
NSF Logo
AT&T Logo
Amazon Logo