Barry Smith explains scientific ontologies in the sense that a scientific ontology always refers to something existing in reality. In the talk, ontologies will be treated on word-level, i.e. neither technology nor implementation are subject of the talk, but only what comes before that in ontological design.