The goal of our research is to develop techniques and tools that, for the first time ever, will allow practical control of privacy of information. The work will allow importing of mobile code, for example, in an active network, while ensuring both the privacy of information belonging to the imported code, and the privacy of information in the node that imports the code.
Our approach will lead to an innovative security model that allows static checking of security properties, a new annotation language for expressing security properties statically, extensions to JAVA that allow code to use the new model, and lightweight tools for checking security properties of both source code (via a new compiler) and bytecodes (via a new bytecode verifier). We also plan to study the runtime support needed by the model, and in particular what is needed to provide a trusted execution platform that runs imported code on imported data while ensuring the privacy of both local and imported information.
We will also develop technology that ensures data integrity in the presence of malicious attacks. This research will define BFT, a new replication algorithm that can withstand Byzantine failures in an asynchronous environment such as the Internet. In addition, the algorithm will be made available via a program library, allowing arbitrary applications to be hardened against malicious attacks. More information on this work can be found on the BFT page.
Andrew Myers leads further research in this area at Cornell University.
Professor Barbara Liskov