Download PDFIn this paper we present the design of an audio conference tool, which is one of our efforts to explore application designs on top of Named Data Networking. Instead of relying on centralized services as current implementations do, ACT takes a named data approach to discover ongoing conferences as well as speakers in each conference, and to fetch voice data from individual speakers. The resulting design is completely distributed and robust against component failures. We discuss design alternatives and tradeoffs, scalability and security considerations, as well as remaining issues for future work.