Authors Singh, Belloni, and Christian demonstrate how visualizations can help students learn some of the most difficult and counterintuitive principles in the physics curriculum. But as two surveys have shown, there are broader roles for computation in that curriculum that ought to be, but currently are generally not being, used to help prepare physics students for their likely work environments.
An August 2002 survey by the American Institute of Physics (available at http://aip.org/statistics/trends/reports/bachplus5.pdf) looked at physics bachelor graduates in the nonacademic workplace at least five years beyond their graduation. The results revealed a significant gap between their computational preparation as undergraduates and the computational demands of their work. The AIP survey does not detail these demands, but from my own experiences in engineering research and development environments, I've found that they include constructing and validating numerical models as well as interpreting results from running those models. In short, holders of physics bachelor's degrees must be able to think about their physics in computational terms.
The other survey, completed by Robert Fuller from the University of Nebraska–Lincoln, provides some answers to how much computation is included in today's physics curricula of colleges and universities nationwide. The answers indicate wide variability in the degree of computation amid a widespread agreement by faculty on the importance of integrating computation into their courses. Fuller concludes that physics departments in the US generally acknowledge the need for more computation in their curricula, but most are not meeting the need in a systematic way. This gap—between acknowledged need and community response—is consistent with AIP's survey findings. The September/October 2006 issue of Computing in Science and Engineering gives Fuller's report and provides some examples of possible ways to close the gap. They include the “lone wolf” who is the sole interested person in the department; the “persuasive pioneer,” implementer of a full computational physics undergraduate major; and a range of cases in between.
I believe the physics community needs to reconceive the canon of the undergraduate physics curriculum to include a significant role for computation. Whether or not they learn their physics principles with computation embedded, students will need to put their knowledge to productive use in their work. Today that usually means through computation.
