This proposal involves the development of an innovative new laboratory for undergraduate and graduate education and research into Internet-related problems. The lab will ensure that our research and students address issues of both immediate and long-term relevance as we move to an Internet-based world. The Internet has changed the face of computer science. Concepts such as e-commerce, grid computing, peer-to-peer programming, and file sharing provide new directions. Old problems, such as security and authentication, have become more complex and at the same time more important. Other problems, such as cooperative work, databases, and distributed systems, have changed scale so dramatically that one now talks of millions of concurrent users. It is difficult to study or teach this new face of computer science using old tools and environments. Many of the complexities and problems occur only when they are tackled at "Internet scale", that is, when there are thousands of simultaneous users, millions of potential users, widely distributed systems, multiple servers, unreliable networks, strong security and privacy concerns, hackers, and the like. Modern research must be driven by and shown to work in the real world on tomorrow's problems. Moreover, students can encounter and appreciate the difficulties inherent in modern Internet-based systems only by working with real systems. The proposed laboratory for experimenting with Internet-scale applications will be suitable for both state- of-the-art research and instruction. For research, it will provide a framework in which ideas can be tested at Internet scale. Such a laboratory will give researchers insights into what the real problems are and how they might be addressed, and will also allow them to validate new approaches. For teaching, the laboratory will show students the problems that arise and techniques used when building applications at Internet scales and.will let us offer courses that better prepare students for real-world programming and research. The laboratory will be used in a variety of research projects. Pervasive programming involves building a common programming framework for dealing with web services, peer-to-peer computing, and grid computing in an open-source environment. Aurora* tackles the problems of large-scale network and stream- based data management. Language research addresses the semantics of web applications. Covalent objects provide a means for object sharing at Internet scales. Prooflets are a simple and efficient way to authenticate large numbers of small data items across the web. In addition, we will use the facility as a laboratory in both undergraduate and graduate courses. Broader Impact: The laboratory will have an impact beyond the immediate research and teaching needs of our department. We will use the laboratory for attracting students to computer science by offering it as a vehicle for both undergraduate and graduate research projects and by making it the focal point of new entry-level courses. The laboratory will be used by a variety of outreach projects at Brown including Artemis, a program for middle-school girls, and TeachScheme!, a program for high-school teachers. The laboratory will also serve as a bridge between academia and industry, providing a framework whereby academics can study and work on industrial-scale Internet problems. Finally, we hope the laboratory will serve as a model for future computer science education and research and we will actively encourage its use by outside researchers.
