Sharing heterogeneous data among distributed enviro nments in a user-centric way represents today the main challenge for personalization. In recent years several techniques have been proposed to support user modeling for multi-application personalization . In this paper we describe G-Profile, our multi- application user modeling system. G-Profile represe nts a way to address some open issues not sufficiently investigated in literature, as the opportunity for a multi-application profile to evolve over the time, together with the possibility to guarantee se curity and privacy in the diffusion of user informa tion among applications. In particular, this paper focus es on user profile data modifications propagation, aiming to reduce the amount of incoherent information about the user over several applications.