Given the various articulated viewpoints on the finer points of ontology design and analysis, I'm curious as to how the group approaches testing, operational validation, and usability testing of ontologies. In software we have unit tests, usability tests, as well as developmental and operational testing of systems. What is the equivalent in knowledge representation? How do we test coverage, usability, etc? How do we develop ontological use cases and how do we validate our ontologies against these tests? I'd be curious to hear opinions from the academic to the operational.