Issue No. 04 - July/August (2006 vol. 8)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MCSE.2006.56
Norman Chonacky , Yale University
I've promised you that CiSE would be broadening its scope during my term as editor in chief, and this issue marks the start of a new chapter in this effort. Sometimes, in seeking the new, we re-encounter the old and are surprised by what we've learned:
We shall not cease from explorationAnd the end of all our exploringWill be to arrive where we startedAnd know the place for the first time.— Four Quartets: Little Gidding, T.S. Elliott
I spent a significant portion of my career as a physics professor. This was sufficiently long ago that I can claim to have been present at the beginning—the entry of computing into the physics teaching enterprise. In fact, this was less than a generation from the time that computers were first used in physics, thus it wasn't entirely clear what the full scope and type of their applications would be. But it was fairly clear that they were surprisingly useful and their application was going to be fairly broad. What I don't think any of us expected was the variety of tasks to which computers would be put nor the depth of the changes they would effect upon the ways we think about the physical world.
Today, a generation or so later, I've found occasion to revisit this domain and be surprised all over again. The occasion is a decision I made as editor to make CiSE more relevant and useful to the physics education community. The surprise I found was how little had changed in the way physics is taught, even though enormous changes have occurred in the way physics—and other sciences, to say nothing of engineering—is done. Computational physics has actually become accepted as a "legitimate" mode of scientific investigation in its own right.
CiSE's association with physics education sprang from the magazine's birth as the result of the marriage between IEEE Computational Science and Engineering (CSE) and the American Institute of Physics' Computers in Physics (CIP). The physics educational community was heavily invested in CIP, and at the time of the merger, CiSE started with a significant number of department editors drawn from that community. CiSE's content, like that of CIP before it, was deemed useful to those teaching physics.
As time progressed, the interesting areas in scientific computing moved from the "garage" to the "factory," and those with a major career investment in research moved along with this trend. But the way physics was taught, and I mean content rather than pedagogy, continued along pretty much the same. Clever new programming methods and algorithms were no longer needed as programming tools got more sophisticated. Ironically, this sophistication required more constant use than many instructors could afford to maintain their proficiency. At the same time, stand-alone products were no longer as easy to create as operating systems became more complex, less transparent, and harder to work within. This meant that many educational users turned to boxed computational applications for instruction and some turned away entirely from trying to include serious computation in their courses. Consequently, computation stayed at the margins in mainstream physics curricular content, and CiSE, now looking more like the old CSE than the old CIP, lost its usefulness for many physics instructors.
Last year, I decided to make revisiting this venue a priority. My first move was to convene a group of physics professors for whom computing hadn't been marginalized, the purpose being an informal conversation to outline computation issues in the undergraduate physics realm. This group had lots of ideas as well as lots of gripes and some laments about the state of computing in the curriculum. One professor, very prominent in this area, declared that the standard introductory physics course content used methodology characteristic of the late 19th century—as if computing hadn't happened at all.
From this meeting emerged the concept for a project and a possible role for CiSE in addressing these issues. The project was to build a community of common interest in the concept of bringing physics practice and physics educational content closer together through computing. CiSE's role would be to enlarge our scope and presentation to more intentionally include the needs of this group of educators for the purposes of facilitating their projects and increasing their numbers.
This current issue is the first step in that direction. In it, you will find some of the usual—articles on state-of-the-art computational science topics such as neural networks and spectral analysis. But you'll also find an article describing how software projects are being used in graduate training for computer scientists and software engineers. Another article hints at the US National Science Digital Library's ability to foster computational physics education. We're also experimenting with our article formats by introducing our first Technical Note, a peer-reviewed article of smaller size and narrower content than our normal feature articles. This one happens to be a clever educational application that makes maximal use of modest resources, and it comes from a physics instructor laboring under stressful conditions: he's from a university in Baghdad, Iraq.
In the next issue, we'll take a special look at computation in physics courses, and it will be the theme of the entire issue. Don't miss it.