This paper is a critical analysis of the concept of ontology thus as it is used in computing science. It identifies three main problems with such a concept, two of which are intrinsic to it and one of which is extrinsic, so to speak, being related to the use of ontology in applications. The first problem with ontology is that the only accepted definition of its main artifact is teleological rather than structural as it would be proper in computing science. The second problem is that claiming that ontology is in any way a semantic discipline requires such a limited and outdated notion of semantic to be to all practical purposes useless. The third and final problem is that the limitations and misconceptions of ontology might make it a limiting factor, rather than a help, for many of the applications for which it is sought. The article concludes that a profound reconsideration of the relation between computers and semantics might be overdue.