This is a pivotal moment for research supporting network-based information management and utilization, as the digital library initiatives come to an end, and new topics such as homeland security emerge to dominate much of the national debate over future information environments and the government's role in creating them. In this workshop, we are bringing together a group of leading researchers in networked information to develop directions and recommendations for long term NSF research.

The intent is to build on the achievements in digital library research and on the priority directions advocated in the recent Cyber infrastructure report (and related studies) to build a long term research agenda. Rather than define another 4-year initiative along the lines of DLI or DLI-2, we have in mind a long term program of sustained research. Ten year goals would not be inappropriate to consider.

In preparation for the workshop, we are asking each participant to write a short paper (3 - 5 pages) that anticipates the content of the final report. You may optionally confine your commentary to specific areas you view as essential. We are asking each of you to imagine that you are personally responsible for defining [part of] the long term agenda for NSF in post-DL research, and for advocating it (i.e., defending the ideas proposed) to the NSF director. We encourage you to email your paper to both Ron and Howard by May 31. Beyond considering a technology roadmap, we implore you to be both visionary and imaginative, while still practical. It is vital to reflect upon the evolving global reach, ubiquity and economy of information and the implication for international projects; to consider both scientific and societal impact; and to be synergistic with the broader national research agenda.

It is our intent to take all of the submittals and to discern overarching themes, directions, and research topics. We will then provide these to the participants prior to the actual workshop, and use the input as guidance in developing the specific agenda for the workshop. The more we can accomplish prior to arriving in Cape Cod, the more successful our discussions will be while there.

While NSF is not DARPA, the 5 questions that DARPA uses to guide program development may help. Attributed to George Heilmeyer, they are:

• What are we trying to do? What is the problem we're trying to solve?
• How is it done today, and what are the limitations of current practice?
• What is new in our approach / technology, and why do we think it will be successful? What gives evidence that it will work?
• Assuming we are successful, what difference does it make?
• How long will it take, how much will it cost, and what are (measurable) milestones, mid-term and final exams?

NSF is indebted to you for your participation, and we realize that we are asking a lot of you. But we do this without apology, for each of you has distinguished yourself as a national luminary in these important areas, and your advice at this crucial time is critical to the future of this research.