Hi all, I was reading this thread on skolems and blank nodes in RDF on answers.knowledgegraph.tech, but the thread is somewhat old. It seems like most of the issues with blank nodes is that they are treated as existential variables, not constants, like IRIs. However, they seem to be the only way to correctly apply a "local scope". Is that correct or is there a more appropriate way of applying a local scope?
In particular, I am interested in building an RDF graph on a single machine without the capability of minting global identifiers, but am trying to utilize the great capabilities of ontologies and SPARQL. Some options I am considering but do not like:
- 1.
IRI scheme urn:uuid doesn't seem appropriate because a single machine should have a way of deterministically assigning identifiers
- 2.
An IRI like myLocalIdentifier:1 would be good because it enforces the resource is a constant, but it doesn't provide any indication that some remote application should treat this particular resource name differently (because it is only locally unique)