We consider a realistic podcast distribution application in remote rural areas, where programs have to be recorded into a CD and distributed to the local radio stations. We use store-carry-and-forward approach, based on mobile agents, that is designed to operate in areas that lack network infrastructure. PrivHab is a georouting protocol that learns the mobility habits of the nodes of the network, then, it uses this information to select itineraries for the agents carrying the data. PrivHab makes use of secure multi-party computation techniques to preserve nodes' privacy. The PrivHab protocol is compared with a set of delay-tolerant routing algorithms and shown to outperform them.