Relational database systems entered the marketplace in the early 1980s and have since become the standard database model. The main reason for this success was the general agreement that they provided a satisfactory re­sponse to the typical needs of business applications. The most successful features of the relational model was the ease of use of its query language, which is set-oriented in nature, together with the conceptual simplicity of the data model. This success has stimulated the extension of the database technology to areas different from business applications, including computer-aided design, computer-aided software engineering, and knowledge representation. These new applications highlighted two major shortcomings of the relational paradigm: the inability to express complex data structures in a natural way, and a limited expressive power of query languages. To overcome the first limitation, the relational model was extended to involve nested relations, thereby relaxing the first-normal-form. To cope with the second limitation, query languages were extended to include more powerful operators.