I've been at the Formal Ontologies Meet Industry workshop lask week. The aim of the forum was to build a bridge between academic research and industrial technology transfer, within the theme of formal ontologies.
I come to the conclusion that there is currently a very small intersection (if any) between the intended meaning of 'ontology' in academia and industry. What industry calls ontologies are in most cases a sort of lightweight schemas used to organize semi-structured data, often something like taxonomies with attributes, with the goal of performing tasks such as querying or data normalization (in particular, no general purpose or terminological "reasoning" using DL-like formalisms). Meaning negotiation is somehow on the horizon, but it usually restricted to a simple usage of W3C "semantic web" standards (RDF, OWL), even when there isn't a real "networked" context.
Academic research, on the other side, is more concerned with foundational issues, like searching a proper axiomatization of time, space, artifact functions, and so on.