Date/Time
Date(s) - 12/08/2015
4:00 pm - 5:00 pm
Location
School of Computing Science
Abstract:
Bad things can happen when online data-sharing systems are in the hands of a single authority. Wikipedia, the product of millions of man-hours, could potentially suddenly disappear if the Wikimedia Foundation runs out of funds. Google and Facebook respectively use algorithms to rank search results and events in feeds that might be biased to maximize ad revenue at the expense of relevance to the user. In this talk, I investigate a paradigm we call Decentralized Social Data-Sharing (DS)^2, which addresses such issues by maximizing the autonomy of data-sharing participants and decoupling the addressing of data from its location. In particular I look at the effectiveness of data queries in such a system, the feasibility of building (DS)^2 applications and whether autonomous peers have an incentive to cooperate and contribute in this setting.
