F-UJI is a tool based on REST to assess of the FAIRness of research data. It has been applied to pilot the FAIR assessment of research data from 5 certified data repositories (PANGAEA, Phaidra-Italy, CSIRO Data Access Portal, WDCC Climate, DataverseNO); these repositories were selected for in-depth collaboration with the FAIRsFAIR project. At present, F-UJI supports an automated assessment of research datasets based on 14 out of 15 core FAIR object assessment metrics (https://doi.org/10.5281/zenodo.3934401) developed by the project. The metrics correspond to a part of or the whole of a FAIR principle. The tool implementation follows the existing Web standards and PID resolution services best practices for research data, F-UJI utilizes external registries and resources to enable programmatic assessment of a data object, such as re3data1 and Datacite2 APIs, SPDX License List3, RDA Metadata Standards Catalog4, Linked Open Vocabularies (LOV)5, and more.

The F-UJI tool is live and already in use. This work is being led by PANGAEA at the University of Bremen. In addition to a free tool, F-UJI source code is Open Source and available within GitHub for co-development and/or forking and re-use. The code is published under a very inclusive MIT licence which allows a broad range of applications including commercial use. F-UJI’s GitHub repository is also linked to Zenodo, which allows the assignment of DOIs to the software. This enables citation of the product and ensures credits to the developers. This will facilitate the spread of the tool within the scientific community, and assist in establishing the tool as a core part of the EOSC FAIR ecosystem. It is anticipated that the accessibility of the Open Source code will assist in the emergence of an active user and developer community. 

After the end of the FAIRsFAIR project, PANGAEA has committed to the long-term hosting and updating of the F-UJI tool. This will include continuing to coordinate the Open Source development of F-UJI’s program code via GitHub and contributing to Open Source community development efforts. The long-term maintenance will also involve monitoring the development of alternative FAIR data assessment tools, such as those developed by other research groups. 

A broad user community currently is forking the Github repository or otherwise contributing to F-UJI via the GitHub issue ticket system. Some are already using F-UJI for FAIR assessments whilst others are evaluating F-UJI for their own purposes. This community includes some important EOSC players, such as EOSCNordic, EOSCSynergy, ARCHIVER, and the European Commission DG for Research and Innovation – European Research Data Landscape project. The main developers located in Europe and Australia have also been promoting F-UJI globally in parallel to the FAIRsFAIR project.

A number of conferences and workshops (e.g. AGU2020, FAIR convergence symposium, STM Research Data Year) have promoted the tool within user communities, while also providing the opportunity to hold focus groups to test and contribute to the tool. Further, two scientific publications have already been published that demonstrate the potential of the tool to increase FAIRness of research data. Further publications are currently prepared and will promote sustainability by involving the scientific community as users as well as contributors to the tool.

While the F-UJI tool is already well-established within the FAIR ecosystem and has clear plans for sustainability, more can be done to enhance the impact of this output. In particular, the development of legal documents (service level agreements, terms of use etc.) will serve as a valuable resource for any further commercial or integrated EOSC activities. Moreover, the development of a business model for longer term sustainability will be helpful for the planning of post-project hosting. The EOSC Association Task Forces (such as the task force on FAIR metrics) are well-placed to lead these discussions, and preliminary discussions have already taken place. It is important to note that a number of funders have already expressed interest in the tool. Indeed, an MoU exists between FAIRsFAIR and the Wellcome Trust to support the FAIRAware project. Similarly, commercial repositories, such as Figshare, have expressed interest in assessing and demonstrating the FAIRness of their data holdings. These existing links will be helpful in future discussions on sustainability and long-term financial support.

Key to the long-term sustainability of F-UJI will be its integration into the EOSC service catalogue and computational environment, including service level agreements with key stakeholders such as EGI or other EOSC core service providers. This will assist with the deployment of the tool on EOSC platforms, as well as its integration with EOSC authentication and authorization systems. It will also ensure that the tool does not contend with sustainability challenges such as vendor locking in the future. Such efforts will be further enhanced by the coordinated uptake of the tool by national as well as regional data infrastructure initiatives, such as the Flemish ECOOM and German NFDI initiatives. This will maximise the exposure of researchers to the tool and support its use in training initiatives. It is also possible that FAIR scores and badges can motivate the creation of an optimised FAIR status for repositories.

A final consideration for the F-UJI tool is the role it could play as a data producer. As the number of F-UJI assessment outputs grows, this dataset could prove a valuable resource for further understanding key areas for FAIR support and training. How such data could be collated and re-used is a topic that the existing user community could take forward after the end of the project.

Source code to the REST API implementation is openly available on GitHub for interested parties to run on their own servers. The same input will replicate to the same output, give the same version of the software is used.