Software has become omnipresent in research as the research process has become increasingly digital, ranging from system software, middleware, libraries, to science gateways and web portals, to computational modeling and data analysis applications. In fact, research is becoming dependent on maintaining and advancing a wide variety of software. However, software development, production, and maintenance are people-intensive, software lifetimes are long vs. hardware, and the value of software is often underappreciated. Additionally, because software is not a one-time effort, it must be sustained, meaning that it must be continually updated to work with in changing environments and to solve changing problems.
Thus, a challenge to the research community is how to sustain software. Tied to this challenge is the fact that in academia, research software may be developed by faculty members, students, postdocs, and staff members, none of whom are generally measured on their software contributions. Can we improve this situation, perhaps by recognizing that developing software can be a creative research process, and that software can be a research output?
While software is increasing important to research, the sociocultural system under which research is performed has not been changing as quickly. This talk will discuss the current state of research software, a number of the challenges around it that exist in academia, and some potential solutions. For example, led by highly motivated research software creators, new career paths and new models for software credit, which have the potential to promote and reward academic research software development, are now developing.