Closed riannella closed 3 years ago
Creating purely inverse relationships is often frowned upon, with the argument that the same effect can be gained by including a single triple in the original direction in the dataset, so having the inverse is unnecessary.
@riannella , is @dr-shorthair 's reply addressing your comment?
If this is not the case, could you please explain the use case behind the proposal of supporting the inverse of dcat:dataset
?
The use case is Aviation datasets. We are working on a mapping to use DCAT now: https://ext.eurocontrol.int/aixm_confluence/display/ADS/DCAT2+and+ISO+19115+representation
You can see there is a requirement in the Dataset to refer to which Dataset Series (dcat:Catalog) it belongs to.
We currently use the general dct:isPartOf property.
(We prefer not to expose dataset users to owl:inverseOf concepts, when we can make the property more explicit)
Thanks, @riannella . So, basically, this is about how to map to DCAT the ISO 19115 element MD_DataIdentification.citation.CI_Citation.series
.
Actually, your approach (dct:isPartOf
) is the same used in the dataset series section of the current DCAT3 ED. Do you think a more specific property is needed?
On a different note:
I see that for the mapping between ISO 19115, DCAT, and Schema.org you are referring to:
https://www.w3.org/2015/spatial/wiki/ISO_19115_-_DCAT_-_Schema.org_mapping
Please note that that wiki space has been frozen, and therefore the mappings have not been updated wrt the latest versions of DCAT and Schema.org.
An up-to-date and more fine-grained mapping from ISO 19115 to DCAT is defined in GeoDCAT-AP:
Hi @andrea-perego - given what is said in Section 11.1 - does this mean using dcat:Catalog as the "data set series" is not recommended?
@riannella , @dr-shorthair , as the discussion has moved to https://github.com/w3c/dxwg/issues/1272 , I'm closing this issue.
It would also be useful to add a new explicit association from Dataset to Catalog. (inverse of dcat:dataset)