Model engineering is taking a growing place in software development. To cope with the diversity of models, the OMG (Object Management Group) itself is now centering its activities on the emerging MOF (Meta-Object Facility) defined for model interoperability. After discussing how the concept of a MOF has progressively emerged in the last ten years, we illustrate some advantages and limits of such a modeling language. We show that explicit definitions of model and meta-model are lacking from the MOF specification and that a linking mechanism from a model to its meta-model is missing. We present the key role played by these concepts and their relations and we propose to define them explicitly within an improved meta-modeling framework. This proposition is based on a proprietary research platform which main contribution is to emphasize the central role played by the meta relation. Making this relation explicit in a true reflective modeling scheme allows us to present a much clearer structure of the overall modeling architecture. © Springer-Verlag Berlin Heidelberg 2000.
CITATION STYLE
Bézivin, J., & Lemesle, R. (2000). Towards a true reflective modeling scheme. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1826 LNCS, pp. 21–38). Springer Verlag. https://doi.org/10.1007/3-540-45046-7_2
Mendeley helps you to discover research relevant for your work.