Web content is now generated by you, me, our friends and millions of other people we do not know. Social networks and collaborative tagging systems have taken off at an unexpected scale and speed (Facebook, YouTube, Flickr, Last.fm, Delicious, etc). This represents a revolution in usage and a great opportunity to leverage the collaborative knowledge to enhance the user search experience. In Gossple, affinities between users are captured automatically, so that users are connected to unknown users, yet sharing similar interests, or exhibiting similar behaviours on the Web. This creates opporunities to fully personalize the search process and increase the capacity of a user to find relevant content. For several reasons, personalisation goes with decentralization: (1) Centralized servers might dissuade users from generating new content and exposing their privacy (2) The amount of information to store grows exponentially with the size of the system and centralized systems cannot sustain storing additional information on a user basis. We believe that the salvation can only come from a fully user centric approach where every participant stores and controls not only her own profile but also her perspective on what portion of the network is relevant to her own search. In tihs talk, I will present the Gossple system and preliminary results on some of its features.