Dear Reader,

Does the project have to die just because the person who started it walks away?

Another angle on this problem is the question of scientific software written by research professionals. Some of the most sophisticated software in the world is created by doctoral candidates and other academic scientists. This software isn't written just to be software – it is written to test a new idea or answer a question related to a research project. Some of these programs represent years of work, but what happens when the developer graduates or gets a tenure track job? Or when the grant used to fund the research expires? More often than not, the project just stops in its tracks and slowly disappears, while the developer seeks new projects and new funding to study other questions.

Academic science is focused on journal articles, not software. The software is a means to an end, so many useful programs are abandoned, and researchers end up reinventing the wheel. Don't ask the PhDs and PhD candidates to solve this problem. No one has ever gotten a distinguished chair for maintaining already-existing software that only a few experts can even understand.

The prospects for orphaned scientific software have become a little brighter with a recent announcement from the Virtual Institute for Scientific Software (VISS) [1]. VISS, which is supported by Schmidt Futures [2], a nonprofit organization founded by former Google CEO Eric Schmidt and his wife Wendy Schmidt, is launching four software development centers at the University of Cambridge, the University of Washington Seattle, Georgia Institute of Technology, and Johns Hopkins University. These centers, which will each employ five to seven software developers, will provide development and support services for scientific projects. Initially, the centers will only work on projects associated with Schmidt Futures, but the hope is to extend that support to other worthy research.

In addition to keeping the software alive after grants end and participants move on, the centers will provide assistance in the initial development phase. Another important, but perhaps less tangible, goal will be to build the identity of the professional academic programmer. Hundreds of professional developers are working right now at universities around the world, but they are often isolated, scattered across the campus, and working independently through separate, unrelated grants. The VISS centers offer the possibility for a collective experience, with the kind of mentoring, work sharing, and synergy that is an everyday part of software development out in the wild.

According to a recent article in Nature [3], VISS is well aware that it can't compete with Internet giants like Amazon and Google in paying top salaries, but they are confident they can still attract high-quality talent. Many professional developers first became interested in coding through their work in science and engineering, and to some, the chance to work on scientifically relevant projects is more exciting than maxing out their salary potential. (And, to be honest, they will probably still fare pretty well compared to a lot of people hanging around a college campus.)

Given the amount of scientific software out in the world today, the addition of 20-30 coders in four small offices won't change the landscape overnight, but the VISS initiative will help to raise awareness about the need to support scientific programming, and it could offer a prototype of a permanent career path for coders who aspire to play a role in the eternal quest for scientific knowledge.

Joe Casad, Editor in Chief