FAIR data and software

Reuse and verifiability

Reuse and verifiability are the main purposes of having research data available. Access to research data (being as inclusive as possible) and accompanying documentation makes science more efficient and more trustworthy. How research data is being managed and shared, depends a lot on the kind of data (observational, experimental, simulation, derived etc.) and the culture within different disciplines and domains.

As open as possible, as closed as necessary

Archiving is already mandatory at Utrecht University, but making relevant data fully FAIR (Findable, Accessible, Interoperable and Reusable) and also open wherever viable (duly respecting constraints of privacy, sensitivity and intellectual property rights) has many additional advantages and is also required by some funders and journals. The adage “as open as possible, as closed as necessary” is valid here.

Data management planning

In order to open up research data, it is important to prepare for proper sharing and reuse with good research data management (RDM) and writing a Data Management Plan (DMP). Though openness and public availability of data is not a strict requirement for FAIR data, not having to ask, draw up a contract etc. does foster reuse, verifiability and experimentation. In this way, deciding to share data openly increases the chances for early reuse and collaborations. 

Code and software

The idea to share code and software openly is older than the promotion of open access of publications and much older than the more comprehensive notion of open science. Free and open source software fosters widespread adoption, user contributions, and ease of collaboration. This manifests itself in two practices, the use of open source software and the open development and sharing of software/code.

Open source software

Complex and sensitive systems run on open source software. Developing software collaboratively using open platforms with version control systems like Git is standard routine for many programmers and has sparked many innovations and speeded up development. In many ways programmers and software engineers are ahead in collaborative, open and transparent ways of working. Sharing code used in data wrangling and analysis is not only an important element in making research verifiable but also to make it easy for others (including commercial entities) to reuse it and improve on it, both of which amount to considerable time gains for the science community as a whole.

Open Science Fellows FAIR data and software