"FAIRable". Measurable. Creditable?

Since the publication of the FAIR principles [1] in 2016, research data infrastructure has evolved significantly. FAIR has provided an important framework, but FAIR is not synonymous with quality. Reproducibility, factual accuracy, and scientific data quality are deliberately excluded from the FAIR principles [2, 3]. What truly constitutes good data and software quality has not yet been defined by consensus or made systematically measurable. Researchers in the Helmholtz Association explicitly call for data and software publications to be recognised in performance evaluations [4]. This wish is difficult to fulfil without clear quality criteria. The situation is further complicated by the fact that established databases such as Web of Science, Scopus, and even OpenAlex still do not systematically index these publications [5]. Several international initiatives call for rethinking research evaluation, yet without measurable criteria, these ambitions remain difficult to operationalise [6]. The Helmholtz Association addresses this challenge with two quality indicators: one for research data (FAIR-C) and one for research software (FAIR-ST) publications [7]. By combining the FAIR principles with a five-level maturity model, they translate abstract quality criteria into a practical evaluation framework, visualised as a radar plot that shows at a glance how a publication performs in terms of openness, metadata quality, or reusability. The indicators are currently in the testing phase; their mandatory implementation as institutional KPIs will begin in 2029. Initial tests at HZDR reveal two structural findings. First, a repository bias: Datasets automatically receive higher ratings when published in established data repositories such as RODARE [8] than when published in an uncurated repository. The indicators currently capture infrastructure maturity rather than intrinsic data quality, with meaningful variance emerging primarily where human judgement is required. Second, an inventory problem: many institutions lack a systematic record of their data and software publications, makingsystematic quality assessment difficult to initiate. The poster presents the indicators' design and invites the data stewardship community to discuss responsibilities for data collection, approaches to missing inventories, and how investments in data and software quality can ultimately be rewarded. Keywords: FAIR principles, research data quality, research assessment, repositorybias, data stewardship [1] Wilkinson, M. D. et al. 2016. The FAIR Guiding Principles for scientific data management and stewardship. Scientific Data 3 (160018). doi.org/10.1038/sdata.2016.18[2] Mons et al. 2020. The FAIR Principles: First Generation Implementation Choices and Challenges. Data Intelligence 2(1-2). doi.org/10.1162/dint_r_00024[3] Miller et al. 2025. A FAIR Perspective on Data Quality Frameworks. Data 10(9), 136. doi.org/10.3390/data10090136[4] Vleugel, M. et al. 2026. Divergence between Perceived and Desired Criteria for Assessing Researchers. Helmholtz Open Science Office. doi.org/10.5281/zenodo.18944700[5] Torres-Salinas, D. and Arroyo-Machado, W. 2026. The 'Big Three' of Scientific Information: A Comparative Bibliometric Review of Web of Science, Scopus, and OpenAlex. InfluScience Editions. doi.org/10.5281/zenodo.18411229[6] DORA: San Francisco Declaration on Research Assessment, 2012. sfdora.org. CoARA: Coalition for Advancing Research Assessment, 2022. coara.eu. Barcelona Declaration on Open Research Information, 2024. barcelona-declaration.org[7] Genderjahn S. et al. 2026. A Framework for Assessing Research Data and Software Publications – The Helmholtz Quality Indicators. Helmholtz Open Science Office. doi.org/10.48440/os.helmholtz.085[8] Helmholtz-Zentrum Dresden – Rossendorf. 2018. RODARE – Rossendorf Data Repository. re3data.org. doi.org/10.17616/R3BR40