The Model-Driven Architecture initiative of the OMG promotes the idea of transformations in the context of mapping from platform independent to platform specific models. Additionally, the popularity of XML and the wide spread use of XSLT has raised the profile of model transformation as an important technique for computing. In fact, computing may well be moving to a new paradigm in which models are considered first class entities and transformations between them are a major function performed on those models. This paper proposes an approach to defining and implementing model transformations which uses metamodelling patterns to capture the essence of mathematical relations. It shows how these patterns can be used to define the relationship between two different metamodels. A goal of the approach is to enable complete specifications from which tools can be generated. The paper describes implementations of the examples, which have been partially generated from the definitions using a tool generation tool. A number of issues emerge which need to be solved in order to achieve the stated goal; these are discussed.

OCL 2.0 is the newest version of the OMG's constraint language to accompany their suit of Object Oriented modelling languages. The use of OCL as an accompanying constraint and query language to modelling with these languages is essential. As tools are built to support the modelling languages, it is also necessary to implement the OCL. This paper reports our experience of implementing OCL based on the latest version of the OMG's OCL standard. We provide an efficient LALR grammar for parsing the language and describe an architecture that enables the language to be bridged to any OO modelling language. In addition we give feedback on problems and ambiguities discovered in the standard, with some suggested solutions.

No abstract

Abstract. Precise modeling is essential to the success of the OMG's Model Driven Architecture initiative. At the modeling level (M1) OCL allows for the precision needed to write executable models. Can OCL be extended to become a full high-level executable language with side-effects? At the meta-level (M2), queries, views and transformations are subjects that will be vital to the success of the OMG's Model Driven Architecture initiative. Will OCL 2.0 become an essential part of the Queries/Views/Transformations standard and what will be its application areas in industry? Can the features of OCL 2.0 be used in the Model Driven Engineering (MDE) approach? This workshop aims at bringing together people from academia that are expected to report on inspiring ideas for innovative application scenarios and tools, and industrial practitioners, which are expected to provide statements on their view of the future of OCL in the context of MDE.