I've been working with Protege, where the 'individuals' are loaded into the 
same model as my ontologies.  In the real world, however, I wonder whether one 
loads the ontologies into a separate model from the actual graph data.  Any 
best practices here?


Reply via email to