No Free Lunch: Research Software Testing in Teaching
Michael Dorner, Andreas Bauer, Florian Angermeir

TL;DR
Integrating research software testing into teaching improves documentation and reduces dependencies, but faces challenges in code integration and academic incentives, highlighting the need for better support and recognition.
Contribution
This study demonstrates the effects of teaching research software testing on software quality and discusses the challenges and benefits of integrating testing into academic settings.
Findings
Improved documentation and fewer dependencies in research software
Student-developed test suites were elegant but not integrated into research code
Effort required for integration was significant and not incentivized
Abstract
Software is at the core of most scientific discoveries today. Therefore, the quality of research results highly depends on the quality of the research software. Rigorous testing, as we know it from software engineering in the industry, could ensure the quality of the research software but it also requires a substantial effort that is often not rewarded in academia. Therefore, this research explores the effects of research software testing integrated into teaching on research software. In an in-vivo experiment, we integrated the engineering of a test suite for a large-scale network simulation as group projects into a course on software testing at the Blekinge Institute of Technology, Sweden, and qualitatively measured the effects of this integration on the research software. We found that the research software benefited from the integration through substantially improved documentation…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsOnline Learning and Analytics
