On Instance-model Querying and Meta-model Transformation

In today's market environment, change is an integral part of all projects. As such, its proper management is a crucial task when it comes to reducing both time and cost of development. The classical modeling approach can improve the situation up to a proper extent but it is not enough, because the process is usually variable and complex. Therefore it is necessary to introduce different level of abstractions for modeling. Each of these levels should serve at a certain phase for a certain purpose in the process. In the early stages several elements are grouped together and aggregated at the higher level, in an abstract model. Throughout time, granularity becomes smaller while understanding of the concepts becomes clearer and clearer and we need to see already working prototypes. In this situation, the approaches on instance- and meta-modeling techniques promise to bring productivity and efficiency to the process. This paper outlines practices from both approaches. We introduce the approach of using Object Constraint Language (OCL) with a Domain Specific Language (DSL) for instance-level model querying, illustrating this method with some examples. We analyze OCL from the broad perspective discussing its advantages and pointing out some its limitations. Moving to a higher level of abstraction, we also present the usage of Kermeta - an extension to the meta-data language with an action language for specifying semantics and behavior of meta-models. We show how Kermeta provides the possibility of automated meta-model transformations.