Background

Bringing together diverse scientific software communities and institutions

One way to characterize scientific software communities is through their institutions. Three distinct categories of institutions that develop, use and support scientific software are corporations, research labs and universities. Each category tends to have distinct business models and sets of priorities. However, the broad scientific software ecosystem is inter-dependent and complementary, where products from one category are widely used in others. Even more interaction is possible if the scientific software community can come to understand commonalities and differences.

Collegeville 2021 focus on software teams

Originally, we thought would we alternate between software sustainability and developer productivity, but we have evolved our strategy to pick an emerging theme for each year. This year’s workshop theme is software teams.

Very little scientific software is developed by indivdual scientists. Instead, teams with diverse skills collaborate on producing and using software to advance scientific discovery and understanding. Understanding how teams function and how teamwork can be improved represents one of the frontiers in improving the impact of software on science. Software team skills and cultures can vary significantly. A scientific software team will certainly have science domain experts but increasingly has expertise in computer science, mathematics and software engineering. As we increase our focus on software teams, we also see value in including expertise in social and cognitive sciences.

What we mean by a software team

We want to interpret software team broadly. We are interested in topics related to small teams, large teams, and a team of teams. We are interested in teams that produce scientific software for themselves (developer-user teams), teams that produce scientific software for others and teams that use (but might not develop) software in their research.

Goals

The goal of this workshop is to bring together community leaders and practitioners in scientific software from academia, industry and labs for the purpose of identifying the challenges and priority research direction for improving teams for scientific software. We want to explore, characterize and articulate scientific software practices from industry, research labs and academia and identify common requirements and essential differences in these communities.

We believe this workshop to be distinct in its efforts to bring academic, industry and lab community members to explore both the common and unique challenges of these communities.

The first day, July 20, will focus on software team definitions and challenges; the second day on technical strategies for improvement; the third on cultural approaches for improvement.

Artifacts from the workshop will include submitted white papers, recorded presentations, live panels and discussions, recordings of live sessions and blog articles from selected workshop contributions. We hope these artifacts will advance community progress in teams for scientific software, especially toward better usability across industry, lab and academic institutions. We anticipate that the diverse experience of the workshop attendees will lead to a deeper awareness and understanding of how the represented communities can benefit from collaboration, coordination and complementarity.

Organizers

• Mike Heroux (Sandia National Laboratories and St. John’s University)
• Keith Beattie (Lawrence Berkeley Laboratory)
• Hannah Cohoon (University of Texas at Austin)
• Argonne National Laboratory
• Vadim Dyadechko (ExxonMobil Corporation)
• Nur Fadel (Swiss National Supercomputing Center (CSCS))
• Jacob Hemstad (Nvidia Corp.)
• Daniel S. Katz (University of Illinois Urbana-Champaign)
• Sarah Knepper (Intel Corp.)
• Addi Malviya-Thakur (Oak Ridge National Laboratory)
• Lois Curfman McInnes (Argonne National Laboratory)
• Pat Quillen (MathWorks, Inc.)
• Elaine Raybourn (Sandia National Laboratories)
• James Willenbring (Sandia National Laboratories)