The information provided here is aimed at institutions seeking to have records describing their research data outputs published in Research Data Australia (RDA). Individual researchers seeking to publish data descriptions in RDA should liaise with their institutional provider.
For institutions, there are two major steps in the process of providing descriptions of research data collections and associated information:
- Creating the records, whether manually, or by processing information from another source.
- Transferring the records to the RDA Registry (which are then subsequently displayed in Research Data Australia).
ANDS registers the description of collections; ANDS does not store the collection itself. You retain control over access to items in the collection. Any special access considerations can be included in your collection description.
If you need access to the RDA Registry, please refer to Getting an ANDS Online Services account for more information.
Most institutions create records in an institutional repository or data store, ideally drawing on information from connected HR and Research Information Systems, and set up an ongoing feed of records that can be regularly harvested, capturing the latest updates. If this is not possible or desirable, records can be created manually in the RDA Registry manual interface.
The RDA Registry requires records to be in RIF-CS schema for ingest. Records can be harvested in RIF-CS, or in another metadata schema used by your institutional repository where subsequent conversion to RIF-CS has been configured.
This task of generating a RIF-CS XML document from records in an institutional repository, and validating it against the RIF-CS schema, can be made easier using the ANDS RIF-CS Java API.
To learn more about providing a record to ANDS and understand RIF-CS you can practice in the ANDS Demonstration Environment (demo), and talk to your ANDS Outreach Officer.
Transferring records to the RDA Registry
The four methods for transferring records to the RDA Registry from an institutional repository are described below. For detailed instructions on configuring the harvest and applying XSL Transformations, see Data source harvest configuration.
Direct harvest means making a single source metadata file available at a URL that can be accessed by an HTTP GET request by the ANDS Harvester. The source metadata can be in RIF-CS XML, constructed at source. If the source metadata is not in RIF-CS XML, the ANDS harvester can ingest the records if an XSLT for converting the source metadata to RIF-CS XML is made available for the harvest. This method is suitable provided the generated file is small (for example, one or two megabytes).
Setting up an OAI-PMH provider to transfer the files allows more advanced synchronisation of content between your system and Research Data Australia. There are several ways to do this:
- Implementing a RIF-CS OAI-PMH Data Provider (overview)
- Configuring jOAI to support RIF-CS (solution using jOAI)
- OAI-PMH metadata harvesting (details)
- OAI-PMH harvest configuration
Catalogue Service for the Web (CSW) harvest
Metadata of any schema can be harvested via the OGC Catalogue Service for the Web (CSW) interface as long as an XSL Transformation is configured for the harvest within the RDA Registry data source account.
JSON metadata of any format can be harvested via the CKAN API as long as an XSL Transformation is configured for the harvest within the RDA Registry data source account.