Ontologies are currently intensively researched in Life Science and Health as they are a means to assure semantic interoperability of systems. However, there is a divide between researchers approaching the issue from an information science and software engineering perspective, and those taking a philosophical stance.

Unfortunately, most ontologies in biomedicine today suffer from a number of serious defects when assessed in light of their conformity to both terminological and ontological principles. This means that much of the information formulated using such ontologies remains implicit to both human interpreters and software tools. Vital opportunities for enabling access to the information in such systems are thereby wasted. These defects manifest themselves in difficulties encountered when the underlying resources are used in biomedical research. Such defects are destined to raise increasingly serious obstacles to the automatic integration of biomedical information in the future, and thus they present an urgent challenge to research.

The general challenge to be met by ontology work is to bridge the gap between clinical research conclusions and the need to make personal decisions in healthcare, and to bridge the gap between data models evolved separately in the two discrete worlds of healthcare and bioinformatics. The following proposals target key issues for the realisation of these objectives and could prevent many of the problems we discuss in the section on information modelling and reference ontology. We cannot state the detailed background for these recommendations here again. It is already stated in other parts this document and can be found in the work of the OBO Foundry initiative, too.