DOE Report on The Science of Scientific Software Development and Use
Summary
The report is substantial (68 pages) had had over forty contributors with the following principal authors Heroux, Michael A., Bernholdt, David E., McInnes, Lois Curfman, Cary, John R., Katz, Daniel S., Raybourn, Elaine M., and Rouson, Damian. Therefore, we consider this human element essential in our efforts to understand and improve scientific software development and use, including engagement with social, cognitive, and information scientists and others who have not historically been part of the DOE computational research community. Theme 2: We need to address urgent challenges in workforce recruitment and retention in the computing sciences with growth through expanded diversity, stable career paths, and creating a community and culture that attract and retain new generations of scientists. Our ability to acquire staff for current and future teams requires strong and sustained efforts to educate, recruit, and retain a diverse workforce by cultivating a supportive and inclusive culture within the computing sciences. Owing to the broader importance of scientific software, efforts to better understand and improve its development and use should find allies beyond DOE, including practitioners in academia and industry, as well as other government agencies that fund computationally and software-intensive research.