RDA 20 Co-located meeting 20 March 2023

Insights into the practicalities of collaboration, data and code sharing across the globe.

When: Monday 20 March, 08:30 – 12:00 CET.
Where: Gothenburg, Sweden, and remotely

The PARSEC team–from five geographically-dispersed countries–has collaborated for four years with several leading data infrastructure and research tool providers. We invited representatives of these key infrastructures and active researchers (users) to discuss the data challenges they had faced (and maybe overcome) and their visions of the optimum path to the future.
A summary of the visions for the future can be found at the end of this article, and the slides are available here .

We want to thank our panellists who provided their time and knowledge.

Panel A: representatives of our infrastructure partners, Crossref (Martyn Rittman), DataCite (Matt Buys), the EDI (Margaret O’Brien), ORCID (Brian Minihan), Scholix (Rachael Lammey) and the World Data System (David Castles).
Panel B: researchers from the fields of marine conservation (Rodolphe Devillers, IRD), health and life sciences (Romain David, EOSC-Life), social inequalities in health (Laurence Mabile, University of Toulouse), machine learning (Jeaneth Machicao, University of Sao Paulo), artificial intelligence (Pedro Correa, University of Sao Paulo), geosciences (Lesley Wyborn, ANU) and domain variations in data stewardship in Japan (Kazuhiro Hayashi, NISTEP).

The slides can be cited as follows:
Specht, Alison, Minahan, Brian, Lammey, Rachael, Rittman, Martyn, Buys, Matt, O’Brien, Margaret, Castle, David, Devillers, Rodolphe, David, Romain, Mabile, Laurence, Machicao, Jeaneth, Correa, Pedro Pizzigatti, Wyborn, Lesley, Hayashi, Kazuhiro, & Stall, Shelley. (2023, March 25). Insights into the practicalities of collaboration, data and code sharing across the globe. Research Data Alliance 20th Plenary (RDA P20), Gothenburg, Sweden. Zenodo. https://doi.org/10.5281/zenodo.7792396

Some of the goals the panellists identified

Infrastructure partners (Panel A). Many were aiming to further link the PID environment to improve discovery of and trust in data and software products. They felt that existing technologies could be leveraged to improve the availability and reuse of metadata. Data standardisation was a real target for future sustainability. Visibility and discovery could be enhanced with richer and more precise metadata for digital objects. Reuse of data and software would be greatly facilitated by improved metadata, and providing aggregate metadata corpuses might be useful as would one flexible data model. Integrating the data and article publication world was seen as one target to be tackled. Many of these goals could be achieved by better tooling, but value narratives could be developed to encourage a rich and diverse community that recognises the importance of identifiers with rich metadata.  

Panel B. The visions that would help research practice differed for our users, although there was some synchronicity. They particularly wanted support, either through the provision of expertise, tools and funding to engage in open data practices, while also emphasising the importance of incentives and rewards for doing so. The latter could partially be achieved through increased visibility of their output but that was only part of it, as institutions and funders have an important role to play. Greater interoperability of code, standards and data was seen as an important goal for the future across institutions, states and countries. Users felt that community involvement in the development of tools was vital (user-centric and more straightforward approaches were mentioned). A confusing data and code landscape was highlighted that needed to be simplified, aggregated, and generally made more manageable, in scale, in diversity and in the applied standards. More visible, readily available and community-driven guidelines, protocols and standards were seen as being a major step forward.