Citation is Collaboration: Software Recognition in Research and Industry
summary
Analysis of 2011–2024 astronomy publications shows 93% of astronomers use Python while formal mentions peak at only about 10–12% of papers and most appear in acknowledgements, not references. Five libraries (NumPy, Matplotlib, Astropy, SciPy, Pandas) dominate usage, yet downloads-to-citation ratios run roughly 1 citation per 1,000–10,000 downloads. Practical fixes: require clear citation instructions, archive releases with citationCFF and Zenodo DOIs, add software to reference lists, and journals must monitor compliance.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 1:11Motivation for software citation
- 3:58Astronomy case study
- 8:11Citation analysis of astronomy libraries
- 14:30Low citation rates implications
- 15:33Credit as collaboration
- 16:15Recommendations for developers
- 18:47Publishing in JOSS
- 19:48Recommendations for users
- 20:51Recommendations for journals
- 21:54Transparent software credit benefits
- 22:37Q&A
related talks
28mWhy and How to Cite Software (and Get Your Software Cited Too)
42mStories in Code
01h 03mKeynote: Scientific Computing and the Gateway to Open Source
54mKey Challenges In the PyData Ecosystem and How we Can All Make a Difference
37mHow do you measure 'public good' in open source and beyond?
44mKeynote
25mEmpowering weather-satellite data users with FOSS: the Pytroll case
45mValues-based OSS Open Discussion