FAIR is not Fair Enough, Particularly for Software Citation, Availability, or Quality

1. FAIR is not Fair Enough, Particularly for Software Citation, Availability, or Quality Daniel S. Katz, dskatz@illinois.edu, d.katz@ieee.org, @danielskatz Assistant Director for Scientific Software & Applications, NCSA Research Associate Professor, CS, ECE, iSchool with Neil Chue Hong (@npch), U. Edinburgh & SSI

2. FAIR for data • Findable, Accessible, Interoperable, and Reusable • Work “started” in 2014, leading to: • https://www.force11.org/group/fairgroup/fairprinciples • Published in 2016: Wilkinson et al, “The FAIR Guiding Principles for scientific data management and stewardship,” https://doi.org/10.1038/sdata.2016.18 • Further definition, 2017: Mons et al., “Cloudy, increasingly FAIR; revisiting the FAIR Data guiding principles for the European Open Science Cloud,” https://doi.org/10.3233/ISU-170824 • Yet more definition, 2018: Hodson et al., “Turning FAIR data into reality: interim report from the European Commission Expert Group on FAIR data,” https://doi.org/10.5281/zenodo.1285272

3. FAIR for software: Is software data? • Katz et al., 2016, “Software vs. data in the context of citation,” https://doi.org/10.7287/peerj.preprints.2630v1 • Backstory – wanted to say software was different than data in the Software Citation Principles paper, but reviewers wanted evidence, so wrote this 12-author preprint via GitHub • “Software is data, but it is not just data” • “While ‘data’ in computing and information science can refer to anything that can be processed by a computer, software is a special kind of data that can be a creative, executable tool that operates on data.” • Differences • Software is executable, data is not • Data provides evidence, software provides a tool • Software is a creative work, data are facts or observations (different licenses) • Rot is different: data has bit rot, software additionally has software collapse • Software lifetime is generally not as long as data lifetime

4. FAIR for software? • Sounds great! • Findable • Accessible • Interoperable • Reusable • Do these terms mean the same thing for software as for data? • Maybe not because of the differences between software and data • So need to define FAIR principles first, likely differently • Is that enough? • Do we also need to redefine the FAIR terms? • Consider citation, availability, quality ✓ ✓ ✓? ✓

5. FAIR for software citation: what else is needed? • FAIR makes credit for data implicit, not explicit • “There are numerous and diverse stakeholders who stand to benefit from overcoming these obstacles: researchers wanting to share, get credit, and reuse each other’s data and interpretations” • Open source model for software leads to liberal forking, pull requests • Software is much more collaborative than data • Since reuse is often collaboration for software -> credit is more important than for data • Then FAIR needs to encourage credit • Metadata clearly and explicitly include the contributors of the software it describes (maybe within F (findable), but not really related to findable) • Citations should be used to record software use/reuse (not sure where this goes)

6. FAIR for software availability: what else is needed? • Much software is open • Open source as inspired by ideal science • Open science as inspired by open source • Is software being open good? • Or is software source code being released under clear licenses sufficient for software availability? • Perhaps FAIR should encourage software to be open • Open should be identified as the default, likely within A (accessibility)

7. FAIR for software quality: what else is needed? • How to judge software quality? • Internal metrics • External usage • (Peer) Review • Is software quality important? • Or is software usage (R) more important? • Perhaps FAIR should encourage software to be higher quality • Quality should be added to FAIR (FAIRQ?)

8. Conclusions • Software is not just data • FAIR for software is not sufficient, particularly for citation, probably for available, and maybe for quality • For citation, credit needs to be explicitly encouraged • For availability, software should be encouraged to be open • For quality, not sure – maybe an addition for FAIR is needed