Supplementing UML with concepts from ORM

Abstract

The Unified Modeling Language (UML) is useful for designing object-oriented code, but is less suitable for conceptual data analysis. Its process-centric usecases provide an inadequate basis for specifying data-centric class diagrams, and the UML graphical language suffers from incompleteness, inconsistency and unnecessary complexity. For example, multiplicity constraints can lead to unexpected problems when extended to n-ary associations, the constraint primitives are not optimized for orthogonality or expressibility, and the graphical language does not lend itself readily to verbalization and multiple instantiation for validating models with domain experts. This chapter examines some of these defects, and shows how to compensate for them by augmenting UML with concepts and techniques from the Object Role Modeling (ORM) approach. It highlights the potential of “data use cases” for seeding the data model, using verbalization of facts and rules with positive and negative examples to facilitate validation of business rules. The following approaches are suggested as possible ways to exploit the benefits of fact-orientation: use ORM for the conceptual analysis then map the ORM model to UML; use UML supplemented by informal population diagrams and user-defined constraints; enhance the UML metamodel to better support business rules.