SIAM's Fall Agenda: Kudos for Journals, Strengthened Ties in WashingtonNovember 7, 2002
At its October meeting, DOE's advisory committee on Advanced Scientific Computing Research, which is chaired by Margaret Wright (second from right), heard from Juan Meza (left) of Lawrence Berkeley National Laboratory and Alan Laub (right), who runs the DOE SciDAC program. With them are SIAM visitors Tom Manteuffel and Jim Crowley.
Thomas Manteuffel and James Crowley
For years, an ever evolving group of dedicated volunteers have worked to ensure that SIAM journals publish the best papers in applied mathematics and computing. Here in the SIAM office, we hear words of appreciation again and again for the high quality of our journals. This summer, the Institute for Scientific Information, after analyzing data on recent journal paper citations, announced that SIAM Review had the highest impact factor, at 2.51, of the 157 journals in the "Mathematics, Applied" category.
As all SIAM members know, SIREV was redesigned in 1999, with new sections containing articles that reflect the broad interests of SIAM members. SIREV includes articles that are intended for classroom use, traditional survey and review articles, short technical articles, and always-popular book reviews. Based on ISI data, the SIREV impact factor was the highest in recent years for an applied mathematics journal. It is a reflection of the community's citation of papers published since the new format was introduced.
We have some especially dedicated volunteers to thank for SIREV's success---Margaret Wright, of course, founding editor-in-chief of the "new SIREV," and the hard-working section editors: Joe Flaherty (Problems and Techniques), Bob O'Malley (Book Reviews), Bobby Schnabel (Education), and Nick Trefethen (Survey and Review). Nick, who for the last four years has had subscribers eagerly anticipating the journal's Survey and Review papers, will be stepping down at the end of this year. We are happy to report that Randy LeVeque has agreed to succeed Nick---and that the other section editors will continue their outstanding work.
SIREV isn't the only SIAM journal to enjoy lofty citation statistics---five of the top fifteen journals in ISI's most recent "Mathematics, Applied" list are SIAM journals. This continues a long-standing tradition for SIAM journals: They offer top-notch content that, having undergone rigorous peer review, appears in accelerated electronic editions and in print, at low subscription prices.
All this makes SIAM sound like a publisher, and, indeed we are; at the same time, though, SIAM is a membership organization. We'd like to give you a little insight on how we balance these two very different functions.
First, as most of you are aware, the SIAM journal peer-review process is run by purely volunteer labor. Although the office staff provides administrative support to the authors and editors, much of the work of processing papers, rigorously reviewing content, and working with referees and authors is done by busy researchers who are not compensated for their time and effort. SIAM does provide a small stipend for editors-in-chief; paid to their institutions, it covers administrative or secretarial support and incidental office expenses. The associate editors and referees for the thirteen SIAM journals, however, work on a fully volunteer basis, providing a true service to the community and fostering the development of scholarship.
What, then, is the relationship between the SIAM journals and the membership?
To start with, SIAM membership is not a criterion for appointment as a journal editor or for publication of a paper in a SIAM journal. Authors who are SIAM members do not get preferential treatment. The quality, importance, and scope of the paper are the determining issues. Although SIAM does not limit itself to the membership in selecting editors, the scope of the journals matches the interests of the community SIAM serves, and a high percentage of our editors are drawn from the SIAM membership.
SIAM's elected leadership (Council and Board members) and appointed leadership, especially the vice president for publications, have direct control over the journal program. They do not get involved in the detailed decisions of peer review, but the vice president for publications does have final approval over editorial board appointments and the scope of the journals' content as expressed in their editorial policies. The VP for publications, who reports to the Council through the SIAM president, is responsible for looking into any problems that need to be addressed.
A bit of news on that front: Tom Manteuffel, with approval by the Council and Board, has named Greg Kriegsmann vice president for publications. Greg fills the position vacated by Mac Hyman when he became SIAM president-elect. Greg brings a lot of experience to the position, having served two terms (1998-2002) as editor-in-chief of SIAM Journal on Applied Mathematics; Pam Cook has succeeded Greg at SIAP.
SIAM's senior elected leaders also approve the creation of new journals and changes to existing ones. The two new SIAM journals, SIAM Journal on Applied Dynamical Systems and Multi-scale Modeling and Simulation: A SIAM Interdisciplinary Journal, were both approved by the Council and Board after a detailed proposal process and discussions among the creators, SIAM staff, and the vice president for publications.
Turning to SIAM's role as a membership organization, we'd like to tell you about some highlights of a mid-October SIAM visit to Washington. In our group, along with the two of us and Gil Strang, as chair of the SIAM Science Policy Committee, were our Washington advisers Mel Ciment and Mark Marin. Confirmed at both stops on our itinerary---the National Science Foundation and the Department of Energy's Office of Science---was the growing importance of applied mathematics and computing to all of science and engineering.
At DOE, we attended a meeting of the advisory committee to the Office of Science's Advanced Scientific Computing Research program. The feeling among ASCR committee members was that DOE's programs are not as well publicized as the more visible NSF programs. Clearly important to a large subset of the SIAM membership, ASCR is a major player in applied mathematics and CS&E research.
Ray Orbach, head of DOE's Office of Science, spoke about computing for scientific discovery, pointing out that high-end computing---both the machines and the algorithms needed to advance scientific discovery---are the "highest priority" for the Office of Science.
Jack Dongarra reported on recent developments in high-performance computing. Especially interesting to the committee were his comments on the Japanese Earth Simulator, which Dongarra sees as a wake-up call to the research community and the Office of Science.
He reminded the group of a series of reports, beginning with the influential Lax report of 1982, that provided a framework for advancing computational science. "It's not just about hardware," he pointed out, repeating an idea heard often at the meeting. Additional funding should be directed to innovative architectures, hardware technologies, software strategies, and algorithm development that will overcome the limitations of current systems, he said; only then will researchers be able to take on the most challenging scientific problems.
The Office of Science is developing a series of reports on areas of science of interest to DOE, such as nanotechnology, detailing the roles that applied mathematics and computing can play in advancing the frontiers in each area.
Nanoscience, according to Bill McCurdy, who chaired a May 2002 DOE workshop on the subject, poses major challenges both to theoreticians and to modelers. The idea that such researchers even have a role to play in nanoscience, he said, was initially greeted with some skepticism. A science, built on experiment, that studies complex phenomena and involves self-assembly---what could theory and modeling have to contribute?
At least the beginnings of an answer have emerged from the May workshop, in the form of a report titled Theory and Modeling in Nanoscience. Based on the workshop discussions, the report details some of the challenges in bridging the great spans of time and length scales necessary to exploit phenomena occurring at the nanoscale.
The topic of multiscale analysis, along with modeling and simulation in general, arose many times in the advisory committee's discussions. Margaret Wright, who chairs the committee, pointed to challenges for numerical methods posed by multiscale problems: "You just can't do the little scales here and the big scales there." Many of the multiscale phenomena that researchers hope to better understand through modeling and simulation present challenging opportunities for researchers in applied mathematics and computing, who are being called on to produce complex new models and computational methods.
Our discussions at the National Science Foundation, in contrast, focused on past successes and current challenges in mathematics. In meeting with us, Bill Rundell, the new director of the Division of Mathematical Sciences, involved his staff in discussing the current status and expected trends within the mathematical sciences.
Such trends include a recent turnaround in graduate enrollments, and even in the number of undergraduates majoring in mathematics. These encouraging numbers were attributed by DMS staff at least in part to programs instituted or augmented in the past few years, including Research Experiences for Undergraduates (REU) and Vertical Integration of Research and Education in the Mathematical Sciences (VIGRE).
Rundell pointed out that recent interdisciplinary solicitations have drawn a strong response---the division was able to fund only 10% of proposals received in response to a geosciences solicitation, for example. This is evidence of strong interest among the scientific community in such research, as well as an indication that such programs are underfunded.
Rundell told us that his goals in accepting the position at DMS included expanding the role of mathematical computation and increasing interactions with fundamental research in other areas of science, while maintaining a strong and vital core research program. Among the areas in which he anticipates growth is mathematics and the life sciences, in which DMS may sponsor a series of workshops.
A major concern was the fate of the proposed increase in the budget for the mathematical sciences. While NSF had requested a $30 million increase for the mathematical sciences, a Senate committee reduced the increase to $10 million---while increasing the NSF budget overall and providing a much-needed increase to the physical sciences directorate. The House committee chose not to follow the Senate lead in marking up its version of the NSF appropriation. Final resolution of this wildly fluctuating budget increase will come only with the House-Senate conference, this year or early next year.