Several of the external sources we use for vocabularies use and contain opaque identifiers (like BODC in collection identifier, URI; https://vocab.nerc.ac.uk/search_nvs/P06/, mentioned in this issue)
Could we somehow make use of these keys=collection identifiers/URIs/namespace+URI to accommodate non-unique language labels for concepts (like the pup issue in the lifeStage vocabulary)?
We already have keys for both vocabularies and concepts so what I am asking is if these keys - particularly for concepts - could be replaced by the external source unique identifiers, in cases where it makes sense. (although my example refers to a vocabulary in this case)
We would want to enable Spanish speaking users to search for cachorro and get occurrences of both pup and cub occurrences (note there is an issue with the whole hierarchical representation of sub-adult, juvenile etc. we also need to sort out if) while English speaking users should be able to search for pup and not get occurrences of cub as well.
@marcos-lg any thoughts on this? The main challenge I see is that we won't consistently have keys=collection identifiers/URIs/namespace+URI to put in the key place. However, opaque identifiers could better support a multilanguage user platform.
Several of the external sources we use for vocabularies use and contain opaque identifiers (like BODC in
collection identifier, URI;https://vocab.nerc.ac.uk/search_nvs/P06/, mentioned in this issue)Could we somehow make use of these keys=collection identifiers/URIs/namespace+URI to accommodate non-unique language labels for concepts (like the pup issue in the lifeStage vocabulary)?
We already have keys for both vocabularies and concepts so what I am asking is if these keys - particularly for concepts - could be replaced by the external source unique identifiers, in cases where it makes sense. (although my example refers to a vocabulary in this case)
We would want to enable Spanish speaking users to search for
cachorroand get occurrences of bothpupandcuboccurrences (note there is an issue with the whole hierarchical representation of sub-adult, juvenile etc. we also need to sort out if) while English speaking users should be able to search forpupand not get occurrences ofcubas well.@marcos-lg any thoughts on this? The main challenge I see is that we won't consistently have keys=collection identifiers/URIs/namespace+URI to put in the key place. However, opaque identifiers could better support a multilanguage user platform.