CISE Distinguished Lecture: Software Sustainability, Routes to Peer Production, Ecosystem Complexity.

Abstract: A key challenge of science policy is to achieve sustained benefit from scientific grant making. In this presentation I will provide a framework for thinking about sustainability in scientific software projects, based on empirical studies of development and use of software in science. The framework starts by asking: what is it that causes sustainability problems? Over time, software declines in scientific usefulness, driven by four factors: a moving scientific frontier, technological change, friction in building software, friction in using software and, least appreciated, change in the software ecosystem sounding a component. These factors drive a need for work; in response we can try to suppress the drivers, try to reduce the amount of work needed, or attract sufficient resources able to undertake the work needed to sustain scientific usefulness. I will analyze three systems by which projects obtain resources: commercial markets, grant-making, and community based peer-production. I will conclude with recent results from a study into pathways to sustainable peer production in NSF grant funded projects.