Comment

Comment by John Gaynard on October 22, 2010 at 10:53am

This is a very complex question. In artificial intelligence an ontology is usually described as a 'reusable terminological scheme'. In addition to being a KM consultant I am a tutor at the Open University Business School. In the 'Managing Knowledge' course on the MBA we teach that ontologies are "schemes for providing a rigorous description of the concepts, attributes and interrelationships deemed relevant to describing a particular aspect of the world." Both computer science and information science can benefit from reusable "agreed models" to describe particular aspects of the world and to ensure common points of reference between different parties. The difficulty lies in coming up with the "agreed models" and that is what brings us back to the fundamental question, 'What is knowledge in this context?' Although good ontologies can be interpreted by computers (that is the objective of the semantic web), it takes the human mind to come up with them in the first place. I hope this answer helps.