Open ddeboer opened 1 year ago
The harvester can handle both schema.org and DCAT. As we'd chosen DCAT as format for the dataset descriptions 'cache', a schema.org to DCAT conversion has been implemented. If we want to store both schema.org and DCAT we need an additional DCAT to schema.org conversion in the harvester.
Additionally, we need to properly document the fact that dataset descriptions are available in our triple store/ SPARQL endpoint in both formats.
Besides that I see no issues.
Currently, DCAT is our data model for datasets.
It would be nice if clients could query the list of datasets both using DCAT/DCterms/FOAF and Schema.org. One of those clients would be the Network of Terms, which uses Schema.org for its catalog. Also, NDE recommends Schema.org for dataset descriptions, so it might make sense if clients could query the Register using the same vocabulary.
One way to achieve this is to store both a DCAT and a Schema.org representation of the datasets. Redundant, but completely fine in terms of LD.
@coret What do you think?