Data Description Registry Interoperability WG Recommendations (Endorsed)


  Data Description Registry Interoperability WG

Recommendation Title: Data Description Registry Interoperability Model

Impact: Provides researchers a mechanism to connect datasets in various data repositories based on various models such as co-authorship, joint funding, grants, etc. 

Recommendation package DOI:

Group Co-Chairs:

Amir Aryani, Australian National Data Service

Adrian Burton, Australian National Data Service

Data Description Registry Interoperability WG is working on a series of bi-lateral information exchange projects and an open, extensible, and flexible cross-platform research data discovery software solutions. Research Data Switchboard is a collaborative project by the members of the DDRI WG. This project leverages DataCite DOI, ORCID and other persistent identifiers, and uses simple but effective research graph technology to link datasets. This system currently links datasets across the following platforms: Dryad, INSPIREHEP (at CERN), ORCID, Figshare and link Australian research datasets through Research Data Australia - supported by ANDS.

This document describes the outcome of the Data Description Registry Interoperability (DDRI) working group and specification of the interoperability model implemented by the partners in this group. In addition, this document shows the testing of this model through an implementation called Research Data Switchboard, a collaborative project by the participants in this working group. The intended audience is the Research Data Alliance community. 


Group content visibility: 
Use group defaults
PDF icon DDRIOutputSpecification.pdf562.87 KB
Plain text icon AdoptionStatement.txt979 bytes
File dc.xml839 bytes
  • Malcolm Wolski's picture

    Author: Malcolm Wolski

    Date: 16 Feb, 2016

    Why was Neo4j graph database used which has its own query language Cypher chosen rather than something like RDF triple store with SPARQL. Wouldn't a RDF/SPARQL approach make it easier to change the implemented product/platform in future and might have enabled greater interoperability with other data providers in a broader linked open data context (e.g. cultural institutions).

  • Amir Aryani's picture

    Author: Amir Aryani

    Date: 20 Jul, 2016

    Hi Malcolm,  sorry I missed your comment. The main reason for using Neo4j in our implementation was the performance of the platform and availability to the developers who could work with this technology in our team. However, the sample implementation can be achieved using RDF/SPARQL approach. 


submit a comment