1521-9615/08/$25.00 � 2008 IEEE
Published by the IEEE Computer Society
Ave Atque Vale
Norman Chonacky Editor in Chief
  Article Contents  
  Conclusion  
Download Citation
   
Download Content
 
PDFs Require Adobe Acrobat
 
It has been my privilege to have served as the editor in chief of Computing in Science & Engineering magazine for the past four years. On 1 January 2009, Isabel Beichl will assume this post as my replacement.
I leave it to Isabel to introduce herself to you in the next issue, but in this, my last editorial column, I want to reflect a bit on what I refer to as our evolving mission. This was, not coincidentally, the title of my first editorial column. In part, my inspiration was a piece written by this magazine's first editor in chief, George Cybenko. In our very first issue, January/February 1999, in reference to our two merged precursors, he wrote:
CiSE is setting up camp at the confluence of two great intellectual rivers—the physical sciences and the computational sciences. This camp will grow into a town and then a city but only if we learn each other's languages and trade in good faith.
Likewise, my first column carried a quote from the magazine's charter, framing its mission as
supporting and promoting the emerging discipline of computational science and engineering, and fostering the use of computers and computational techniques in scientific research and education [] among physical scientists, engineers, mathematicians, and others who would benefit from computational methodologies.
George's metaphor echoes that charter passage, and both pieces serve as context for my reflections about what I've seen transpire in and around CiSE during my four-year tenure. In particular, as I stated in that first column, my commitment as EIC was to focus on three aspects of our work: community, content, and approach. My intent was to foster community as a marketplace for cross-cutting ideas, to diversify our content as a melting pot for computation in all sciences, and to explore approaches to achieve these ends by experimenting with different venues.
For one, the target community related to our mission has certainly grown in the past four years. Technological advances that have made more computing power available to the "masses" have spurred increases in computing across the sciences and engineering. Multicore processors and cheap, fast storage, now found on any desktop, make computational approaches to doing science accessible to anyone who recognizes how to apply them. Building on new computational power and storage capacity, Google has revolutionized the way users can access information, from searches to maps to collaboration. Improved simulation techniques and tools have greatly expanded the range of scientific and engineering problems that are accessible to solution.
However, despite this expansion in computational capabilities and the conceptual innovations in their application, it's commonly accepted that the rate of growth in computational usage by science and engineering fields—particularly at the high-performance end and among experimentalists—lags. Just as the computational sciences and the sciences of computation are inextricably intertwined, so are the challenges of education, invention, and practice to realizing computational benefits.
I experienced part of this on a personal level in my EIC-related journeys, both geographic and intellectual. In 2005—my first year on the job—I attended a full poster session at a large, national, applied optics conference. Out of the 100-plus posters, I could find only two in which numerical simulation was an integral part of solving the experimental problem. One of the presenters, a graduate student, freely offered that the simulations he used were an important part of unraveling the mechanisms responsible for the experimental data. He also mentioned that it was his initiative and responsibility to set up and employ the simulations (adapted from a national lab's code) even though it was his professor who was responsible for recognizing the results' meaning.
Among many other examples I've witnessed during these four years, this one illustrates paradigmatically the nature of the barriers to greater computational use. Based on them, I will speculate that they're partly generational, partly cognitive, and partly experiential. First, the generational: compared as cohorts, the younger generation simply has more ease and more appetite for engaging computation than the older. second, the cognitive: both cohorts suffer from a relative lack of ability at "computational thinking." Computation truly is a third paradigm—along with experiment and theory—for coming to understand the physical world, and many if not most of us simply aren't up to it yet. Third, the experiential: the community suffers from a lack of successful examples and thus isn't conducive to taking risks that have real payoffs.
This situation is what makes the prospects for publications such as CiSE so tantalizing. situated "at the confluence," as it were, of two enterprises with such great prospects, it's positioned to exploit its role as a magazine to cross-fertilize among our component constituencies. we can, by virtue of the magazine venue, offer case studies of new methods, perspectives, needs, and successful examples in a form that can be widely understood not just as to their details but also in their implications for our own work. This is our holy grail.
So, how have we been doing? That answer is up to you, the readers, to decide.
What about our future? The bottom line is that all paper-based technical publications have considerable costs and are suffering financially. This vulnerability was one of the many things that I had to learn about the publishing enterprise. I had excellent teachers in the professional staff and directors of the IEEE Computer society publications office and many "learning moments." True, the individual subscription level for CiSE bucked the industry's trend to plunge last year, and we actually experienced an increase, but we should be doing much better. One of the movements within the publications enterprise is to understand how to better use information technology to present content and, more important, value-added services. However, such paradigm transitions for publication suffer analogously from obstacles like those I just attributed to computation in science and engineering.
One obvious service that CiSE might provide is tied to the considerable need for building community among computational scientists of all flavors—physical, geophysical, astrophysical, chemical, and biological—and those engaged in scientific computing—computer scientists, applied mathematicians, and software engineers. we've made strides during the past four years in expanding our content's scope over this full spectrum, and our editorial board is becoming correspondingly diversified. But we're just now beginning to experiment with technologies such as wikis for interaction and collaborative work, with much to learn. Moreover, we don't have unlimited time to make a go of this experiment "at the confluence," as George so aptly described it.
Conclusion
Fortunately, I leave this post in able hands, with a committed, working board and a brand-new EIC. I should point out that Isabel is an applied mathematician whereas I'm an experimental physicist. we had an understanding when our predecessor publications merged in 1999 to form CiSE that EICs—each with two two-year terms—would be chosen alternately from the two broad areas represented in our parent publications, and the selection of an applied mathematician honors that agreement. as publications often reflect the professional personality of their EICs, so will CiSE and with good purpose. after all, our charter calls for working together across differences, and we shall continue to do so.
With this, I extend all good wishes to Isabel for the future. Hail and farewell.

Norman Chonacky is interested in hearing any and all thoughts you might have about this magazine. Contact him at cise-editor@aip.org.