Scientific research is increasingly reliant on advanced social and technical infrastructures called cyberinfrastructure (CI). In turn, the cornerstone of scientific cyberinfrastructure is Scientific Cyberinfrastructure Software (SCIS) that enables data collection from digital instruments, analysis of data through the execution of mathematical algorithms, data visualization and sharing of data through standardized file formats and widely accessible databases. Despite the importance of SCIS for data-intensive research, too little is known about how scientists actually use, adopt and develop scientific software. More research is needed to explore how software, software development and software sharing practices are, and can be, community products, resources or practices.
This research project will examine how scientists develop SCIS as part of their day-to-day research practice through a qualitative, ethnographic study of 6 research groups using observations, semi-structured interviews, and document analysis. The researchers will investigate how decisions about SCIS are made; document, classify and analyze actual practices for using, adopting, developing or sharing software; identify scientists' incentives and disincentives to share software at the local, organizational and community levels; and discern the impacts, intentional and unintentional, that SCIS systems have on scientific data and the scientific research process.
Understanding SCIS development and sharing is necessary to ensure continued integrity of datasets shared within and among communities, facilitate the sharing of the tools and practices that are developed using national research funds and most importantly, support a fundamental tenet of scientific research: the open communication of the processes and practices behind published research findings. Furthermore, understanding these practices will support the education and training of our nation's future generations of scientists and engineers. This project will thus benefit domain scientists, SCIS developers, Computer Supported Cooperative Work (CSCW) and Software Engineering (SE) scholars and policy makers by providing conceptual tools that will provide guidance when considering how and when software can be designed and supported to be accessible and useful to the broader research community.