The challenges of theory-software translation

Area of Science:

• Scientific computing
• Computational science
• Research software engineering

Background:

• Software is integral to modern research, requiring it to represent complex theoretical constructs.
• Ensuring the robustness of theory-software translation is critical for scientific integrity.
• Challenges in this translation process have been historically under-explored.

Purpose of the Study:

• To identify and categorize challenges in translating scientific theory into research software.
• To explore the implications of these challenges for scientific advancement.
• To propose a new research area focused on theory-software translation.

Main Methods:

• Thematic analysis of discussion sessions from the Theory-Software Translation Workshop 2019.
• Involved academic researchers and research software engineers from diverse domains.
• Focused on expertise in high-performance computing.

Main Results:

• Identified a broad spectrum of challenges in implementing scientific theories within research software.
• Categorized challenges into emergent themes: design, infrastructure, and culture.
• Mapped identified challenges to specific research questions.

Conclusions:

• Systematic investigation of research software construction and output utilization can enhance robustness.
• Propose the establishment of theory-software translation as a new research area.
• This new field has the potential to advance both scientific knowledge and practice.