This post has been authored by Sarah Reiff Conell (Princeton) and Sarah Wright (Cornell) who both attended the International Digital Curation Conference in February 2024 held in Edinburgh, Scotland.
We recently presented a poster titled ‘Desirable Characteristics and Trust in Repositories: a cross-institutional comparison’ at IDCC 2024 in Edinburgh, Scotland. The poster summarized the self-evaluation performed by DCN member institutions in response to the 2022 National Science and Technology Council report, “Desirable Characteristics of Data Repositories for Federally Funded Research.” The characteristics included a range of features that were thematically grouped: organizational infrastructure, digital object management, and technology. These themes are central to ensuring that data resulting from federally funded research is broadly accessible, robustly curated, and preserved over the long term, and DCN curators met and discussed how well our member institutions were meeting these characteristics.
The 17 institutional, data, and generalist repositories that participated in the conversation generally agreed that they satisfy the desirable characteristics. Some characteristics were identified as having room for improvement, requiring researcher buy-in, or existing in a state of transition or ambiguity. Most DCN member repositories do not take human subjects data, so we did not address those additional considerations.
Many responses in our discussion included the phrase “it depends,” since researcher buy-in plays an important role in completing the metadata and documentation necessary to create and publish FAIR data. We often ask for the information, but many of us don’t require it, thus “it depends” whether the resulting datasets fulfill the characteristics.
We offer example documentation as one strategy for repositories to increase transparency and trust in using them as a solution for sharing and preserving federally funded research. Gaps identified can also help determine where institutions and organizations can invest and improve repository infrastructure and services. Therefore, this documentation can be used for communicating to administrators and institutional leadership as well as to researchers considering depositing into our repositories.
We concluded that we all need to work to create better documentation of the policies and recommendations of our repositories, as well as the role of the researcher in creating high quality and FAIR data deposits.
During the conference, Martin Halbert of the National Science Foundation, who helped draft the characteristics, was pleased to see how they have been used across and discussed by our member institutions.