COAR hosted a members-only webinar on 5 December on role of open repositories in research data management (RDM). RDM is a large component of open science, and the role of repositories is crucial in supporting RDM. Many repository services that began by collecting articles are interested in expanding their operations to include research data but it is difficult to know where to start.

Several COAR members shared their stories about how they began their research data management services. Our speakers Robin Rice from University of Edinburgh, Sven Vlaeminck from ZBW Leibniz Information Centre for Economics, Barbara Hirschmann from ETH Zurich, Lluis Anglada & Mireia Alcalá Ponce de León from Consorci de Serveis Universitaris de Catalunya (CSUC), Rebeca Marin Del Campo & Marisa Pérez Aliende from Madrono Consorcio and Jose Antonio Galeano Cardenas, Del Rosario University greatly contributed to this discussion.

While every organization and jurisdictional context is different, there are many shared challenges with adopting research data management services. Given the complexity and expertise related to RDM, some organizations can find it overwhelming. But, organizations just have to start somewhere, and then build and expand expertise and services incrementally over time.

The research data management lifecycle can work as a good metaphor for designing services: “before research begins”, “while work is in progress”, “at the end of research”. Absolutely critical is that RDM service help researchers. If services are built on and respond to existing needs and requirements of the research community, they can greatly help increase the reputation and visibility of the library as a partner in research.

Key issues for organizations are:

  • How to balance the need to develop trusted services based on best practices vs. the resources required to adhere to stringent standards. Are there some feasible FAIR data practices and certification processes?
  • Identifying the main interoperability requirements for linking data and other types of content and integrating different data sets. Geographic location is one element that works for many datasets, but more recommendations are needed.
  • Defining the collection policy for the repository. Will the repository collect any type of research data produced by affiliated researchers, focus on a specific domain specific, address long tail RDM needs only?
  • Should different platforms be adopted for RDM? While existing IR platforms have extended their functionality to include research data, there have limitations. As services mature, the organization may need to consider systems that specialize in RDM.
  • Preservation services for RDM are still in their infancy. The resource required for long term preservation often are beyond the existing service

We thank the presenters for sharing their experience with the COAR community and COAR will try to help address some of the challenges identified above through its RDM working group.

Webinar recording and the slides of the presentations are now available: