Software is indispensable in quantitative research: it enables you to analyze collected data and turn it into visual representations, such as graphs and tables. Because application requirements vary per study, researchers often write this software themselves—from simple algorithms to complex analysis tools. However, this self-written software is still rarely shared, Linus notes. ‘That’s a shame, because other researchers can really benefit from it. If you share your software, they can reuse or adapt it for their own research.’
Reuse and integrity
In his new role as Research Software Trainer, Linus wants to advise researchers on sharing self-written software. His role is similar to that of the Data Stewards, who have long supported Radboud researchers in making data publicly available. ‘That too is done with a view to reuse and research integrity’, says Linus, who also works part-time as a Data Steward. ‘When sharing data, it's important that it meets the FAIR-principes. In other words: the data should be Findable, Accessible, Interoperable (able to work with other software), and Reusable. These principles also apply when sharing self-written research software. That way, as a researcher, you not only show the data you collected, but also how you analyzed it, maintaining scientific integrity.’
There’s a lot involved in sharing self-written research software according to the FAIR principles. ‘Especially if you want others to be able to reuse it’, Linus explains. ‘First, it’s important to decide on which platform to publish the software and ensure that copyright requirements are met. It's also essential to clearly explain how the software works—what kind of input it needs and what kind of output can be expected. I advise on all these criteria. With reusable research software, you help other scientists take the next step and build upon your work.’
Collaborating and referring
Linus aims to organize various workshops on this topic, focusing on best practices rather than teaching a programming language. ‘Within each faculty, there are colleagues who can assist with programming—I can connect you with them. There are also online learning modules that cover programming, and I can refer you to those as well. And if you have questions about writing understandable, reusable software code, I’d be happy to brainstorm with you.’
In his role as Research Software Trainer, Linus combines his research background and software expertise. He originally started as a researcher in Historical Celtic Linguistics, and in recent years he worked as a product owner and developer at the Donders Institute. As a Radboud researcher, you can reach him by email, phone, or Teams for advice or an in-person meeting on campus. ‘The ideal situation is that researchers take FAIR principles into account from the very start when writing software. That prevents you from having to rewrite things afterward. My goal is to work closely with colleagues in the faculties who have programming expertise, so we can support researchers early on in meeting the accessibility requirements for their self-written research software. Hopefully, together we can build university-wide awareness at Radboud and achieve a cultural upgrade.’