Resolving heterogeneity among information systems is a crucial necessity if we wish to gain value from the many distributed resources available to us. Problems of heterogeneity in hardware, operating systems, and data structures have been widely addressed, but issues of diverse semantics have been handled mainly in an ad-hoc fashion. In this paper, we present ONION, a system based on a scalable approach to interoperation of information systems that articulates their associated ontologies. An articulation focuses on the semantically relevant intersection of information resources with respect to a type of application. However, ontologies obtained from diverse sources are represented using different formats. We have designed a simple intermediate format - the ONION ontology format - that we transform ontologies to before we generate semantic correspondences or articulations between them. In ONION, application-dependent articulation rules that capture the correspondence between concepts in...