Data Fabric IG: Repository Registries

The issue of Repository Registries was discussed with different views at 3 plenaries. The first discussion at P3 in Dublin led to a stable re3data registry setup and the last session at P6 in Paris turned out with a categorisation of three different registries in addition to the typical metadata data registries which are being offered in many communities already:

  • repository registries such as re3data that collect much useful information about repositories for human consumption mainly so that depositors and users can easily find where to go to for their needs
  • collection registries such as proposed in the session where major domain repositories describe their major collections also meant mainly for human consumption informing about typical data collections that can be accessed. It was agreed that such a registry would be a useful appetizer at the front page of the RDA web-site.
  • federation repository registries (FRR) as they are required in the increasingly large federations that are currently in development worldwide and that are meant mainly for machine consumption to enable efficient operation

Currently, all these many registries, which are maintained by service providers, request different types but partly overlapping attributes from the repositories which is creating increasing problems in maintening this information. It was argued that it would be much better to have a standardized schema offering sufficient flexibility so that repositories only need to maintain one description which then can be used by the different service providers to extract the information they need. Since re3data is running making use of a well-defined schema two activities were started:

  1. Setting up a collection registry for easy discovery purposes
  2. Defining a consolidated and harmonized schema based on what is already being used

Therefore, we introduce the following sub-pages:




submit a comment