—Recently, mobile social software has become an active area of research and development. A multitude of systems have been proposed over the past years that try to follow the success of their Internet bound equivalents. Many mobile solutions try to augment the functionality of existing platforms with location awareness. The price for mobility, however, is typically either the lack of the popular friendship exploration features or the costs involved to access a central server required for this functionality. In this paper, we try to address this issue by introducing a decentralized method that is able to explore the social neighborhood of a user by detecting friends of friends. Rather than only exploiting information about the users of the system, the method relies on real friends, and adequately addresses the arising privacy issues. Moreover, we present VENETA, a mobile social networking platform which, among other features, implements our novel friend of friend detection algorithm.