ICSU-WDS & RDA Data Publication Services WG

1. ICSU-WDS & RDAData Publication Services WG

2. Linking Research Data and the Literature: why? Linking Research Data with the Literature is of great value, yet current solutions are not realizing the potential • Why link? • Increase visibility & discoverability of research data (and articles) • Place research data in the right context to enable proper re-use. • Support credit attributionmechanisms • Examples • Some data repositories keep track of articles that cite, or refer to, their data • Some publishers have applications to link articles with data hosted externally • Providers of bibliographic information and infrastructure providers are taking efforts to “connect the dots”

3. So.. what’s the problem? Linking Research Data with the Literature is of great value, yet current solutions are not realizing the potential • What is the problem? • Many disconnected sources (publishers, data centers, repositories, infrastructure providers, …) • Heterogeneity of practices, for example: • Different PID systems (DOI, accession numbers) • Different ways of referencing data (formal citations, in-text references, …) • Different moments of citing data (at publication, post publication, …) technical social

4. It’s all about connecting the dots service

5. What do we propose? Objective: move from a plethora of (mostly) bilateral arrangements to a one-for-all service model infrastructure for the research data publication landscape • Increase interoperability • Decrease systemic inefficiencies • Power new tools and functionalities to the benefit of researchers

6. So.. what will this “orange blob” do? • Given article A, what relevant data D exists – and vice versa • Additional metadata about the nature of the relationship, e.g. supplementary data, related data, formal citation. • Additional metadata for article and/or data set

7. General Recommendations The ideal data / literature interlinking service • Universal: cross-disciplinary, global • Inclusiveand participatory: supported by all stakeholder groups • Openand non-discriminatory • Quality through meticulous provenance and metadata (not “filtering at the gate”) • Standards-based

8. General Recommendations – long-term view • The ideal data / literature interlinking service • The “multi-hub model” • Infrastructure & service layer • Create sustainable infrastructure as extensions of existing systems • “Follow the content”: use established processes as natural aggregation points (“hubs”) for different constituencies • Interoperability between the hubs through common standards • Inclusive – new hubs welcome

9. WG Output: the “Data-Literature Interlinking (DLI)” Service (Prototype) interlinking service developed with OpenAIRE and PANGAEA SearchAPIs OAI-PMH Web Portal OAI-PMH intersection Information Space PID resolving De-duplicating Core Data Model Harmonizing Over 1.4M article/datalinks! • Examples: • Pairs of DOIs • DataCite records • PANGAEA records Links collection … Data Sources

10. WG Output: the “Data-Literature Interlinking (DLI)” Powered by OpenAIRE D-NET software and PANGAEA search engine Give it a spin: http://dliservice.research-infrastructures.eu

11. Examples of adoption & implementation • Long-term view is adopted by CrossRef, DataCite and OpenAIRE – supporting the “multi-hub system” infrastructure • Europe-PMC has adopted the DLI metadata standards • Connected with RD-switchboard (output of RDA WG “Data Description Registry Interoperability”) • Ad-hoc information requests on linked data • Several data repositories are exploring connection with current API interface.