Workshop: The Future of Internet Research; Berkeley, CA; December, 2003

International Computer Science Institute, Berkeley CA

Investigators

Abstract

The Internet has been one of the proudest success stories in computer science research; visionary researchers aided by farsighted funding agencies changed the world in a fundamental, and unexpected, way. However, Internet research (by which we mean networking research focused on the Internet and related networks) is now at a crossroads. The Internet itself is now fully commercial, and less amenable to input from the research community. Many researchers struggle with the choice between evolutionary research (which might influence the infrastructure) and revolutionary research (which will almost definitely not have such impact). Moreover, many outside the field consider the Internet, at least from the networking perspective, to be a solved problem. For those in the field, there is little doubt that there is much more work to be done, but there is little agreement on what that work is. The workshop will focus on three main topics: o Are we done yet? After all the progress that has been made in the last 20 years, we must ask ourselves if there are remaining technical challenges that require our attention. What questions should the field be focusing on? o How can we formalize what we know? While Internet research has created an amazing artifact, it has created little in the way of fundamental scientific foundations; Internet design lore is well accepted, but poorly formalized. Can we create a .science of networking that applies to the Internet and other similar networks? o How do maximize the impact of our research? In the early years, Internet research had a tremendous impact on the deployed infrastructure. Now, much of Internet research has little effect on the commercial offerings. How can we increase the impact of our research? What is the right mix of evolutionary versus revolutionary research? Are there infrastructures, such as overlays or open-source routing code, that would increase the impact?