Title: Distributed Social Networking with Personal-Cloud Butlers
Authors: Seok-Won Seong, Sudheendra Hangal, Debangsu Sengupta, Greg Bayer, Jiwon Seo, Monica S. Lam
Keywords: Social networking, Cloud computing, Mobile devices
Submitted to: Third International Conference on Weblogs and Social Media (ICWSM'09)
Abstract
To make our personal information accessible from anywhere and to share it with our social network, we are storing increasing amounts of data in the cloud. The
network effect is driving the establishment of a few giant corporations that own our private data, possibly even leading to a monopoly. As a result, we face loss of privacy to third-party for-profit organizations, loss of competition in the market place, and increased risks of large-scale fraud. This paper proposes a distributed system design that gives data ownership back to the consumers and offers choice in where we want to store our data. We introduce the concept of a Personal-Cloud Butler, a personal server program that presents a coherent view of data stored in a federation of distributed devices and supports sharing of private data among friends. Central to the system is a PrPl (PRivate-PubLic) semantic index, which is used to index all the data located on different storage devices. With the help of URIs (Uniform Resource Identifiers), the protocol eliminates data copying by allowing applications to retrieve data directly from
storage. By providing active mediation of access to a distributed collection of data, we enable new applications that integrate together information from federated
data stores while honoring data privacy. We have implemented a prototype of the Personal-Cloud Butler and have also developed a collection of applications as a first demonstration of the feasibility of this approach. By establishing an open, distributed, convenient social networking API, we can potentially shift the paradigm by providing better control over our privacy in social networks.