The More Things Change . . .June 12, 2006
Anil Deane, newly appointed program director in the Department of Energy's Mathematical, Information, and Computational Science program.
Talk of the Society
Things change, perhaps nowhere more certainly than in the areas of science policy and science funding. As of November 2005, the picture for science funding in the U.S., especially at the National Science Foundation and in the Department of Energy's Office of Science, appeared somewhat bleak. A few months later, in the State of the Union ad-dress, came the announcement of an initiative to in-crease investments in science and engineering; details became clearer in subsequent days with the release of the President's FY 2007 budget.
After dramatic increases during Rita Colwell's term as director (1998–2004), funding for the mathematical sciences at NSF has been flat. With the unveiling of the President's 2007 budget, it became clear that the mathematical and computational sciences would not fare especially well, even in a budget intended to enhance economic competitiveness.
This was the backdrop for several recent policy meetings in Washington: the Joint Policy Board for Mathematics, an umbrella organization of the AMS, ASA, MAA, and SIAM that focuses on policy issues of common interest (April 24) and the SIAM Committee on Science Policy (April 25). On April 26, at a meeting with a small group representing the AMS and SIAM, NSF director Arden Bement expressed the view that mathematics and statistics, along with computational science, need to be recognized as important elements in the strategy for implementing the President's proposed increases for science.
As usual, SIAM's science policy commitee had arranged for several guests: program managers in various funding agencies, as well as officials from the White House Office of Management and Budget (OMB) and Office of Science and Technology Policy (OSTP).
In the course of the day, the visitors reported on activities at their agencies, with the emphasis on budgets and future scientific directions. The committee learned that in the President's budget, funding for NSF's Division of Mathematical Sciences would increase by only 3.2%, while that for the parent directorate of DMS (Mathematical and Physical Sciences) would receive a 6.0% increase and NSF's overall research budget would grow by 7.9%.
Another visitor had come from CISE (NSF's Directorate for Computer & Information Science & Engineering), which houses computer science research. Michael Foster, director of the Division of Computing and Communication Foundations (CCF), pointed out that the division is considering a few algorithm-centric thrusts for FY 2008. His main message, however, is one that we have heard before: If we want our voice to be heard, we need outstanding people from the community---people who understand and can present the community's views on a wider stage---to serve as program officers. Indeed, CCF is looking for an outstanding computer scientist with a background in scientific computing or related areas to fill an open position. Taking such a position is one of the most important contributions an individual from our community can make. Another important activity is to provide agency officials with success stories--short, accessible written descriptions of advances in an area that they can use to make the case for the value of research in that area.
Foster closed with a description of GENI (Global Environment for Networking Innovations), a major proposed initiative from CISE. The idea behind GENI is to redesign the Internet from the bottom up and create a shared facility funded through the major equipment (as opposed to research) account. "This will inspire our programs for the next year or two," he said.
Two representatives of DOE's Advanced Scientific Computing Research (ASCR) program brought the committee up to date on program activities. Anil Deane of the Institute for Physical Science and Technology at the University of Maryland, College Park, recently began a term as a Mathematical, Information, and Computational Science (MICS) program director. Senior ASCR technical adviser Dan Hitchcock, a frequent CSP visitor, also engaged in discussions with the committee. Presenting the current budget and future plans for ASCR programs, Hitchcock said that the Office of Science is slated for a 14.1% increase for research in 2007, with substantial increases for ASCR. Much of the increase will be allocated to new hardware acquisitions, with research in algorithms and software that will be needed if these machines are to be used effectively to come in future plans.
The committee applauded the substantial growth of the SciDAC program, which since 2001 has been funding application-specific development of algorithms and software in areas of DOE's mission. At the same time, committee members voiced the opinion that the Office of Science should also support research in the science of computation itself, not directly tied to a specific application; payoffs of such projects have proved equal to or greater than those attributed to improvements in hardware.
OMB budget examiner David Trinkle had encouraging words for the committee: "The argument for mathematics is compelling" in that the mathematical sciences underlie much of economic competitiveness. Nonetheless, he pointed out, it is up to the community to make that case.
Between them, Trinkle and Rob Dimeo of OSTP covered the President's initiative in science. The focus of the initiative, they said, is on innovation that will enable technology. As Trinkle put it: "Where would I look, where can I make an investment that will lead to something?" Committee members responded with views on the role of modeling and simulation, risk analysis, and improvements in computational capability that arise from new algorithmic approaches.
Changes in agency budgets and priorities did not fill the day's agenda entirely: The committee also discussed recent changes in the crucial human component of the agencies. Most significantly, this was the last CSP meeting that Bill Rundell would attend in his capacity as DMS director. Rundell, who has held the position since 2002, steps down this summer. The CSP will miss his candid and sometimes provocative comments, and sincerely thanks him for his dedication and energy.
Succeeding Rundell will be Peter March, chair of mathematics at Ohio State University. The committee looks forward to meeting and working with him in the future.
As mentioned earlier, Anil Deane recently began a term in DOE's MICS program. Michael Strayer is the acting director of MICS, and the associate director of ASCR, where he is the successor of C. Edward Oliver.
The committee welcomes those who have joined the agencies, either in permanent positions or as rotators. An often heard message emerged clearly from the discussions at the CSP meeting: An individual can make a huge impact as a program manager, and the community needs good representatives in the agencies. Some things never change.