MIDST (Model-Independent Data and Schema Translation) has the goal of developing tools for the translation of database schemas and instances from a model to another. The approach is based on a "metamodel" notion: data models are described with reference to a small set of metaconstructs, and translations are specified on metaconstructs as well, so that they are reusable. The project started in its present form in 2003, as a follow-up of a previous project carried out in 1991-1997.
In MIDST, new techniques have been proposed for database translations from a model to another, for example from object oriented to SQL or from SQL to XML schema descriptions.
The approach leverages a predefined, but large and extensible, set of models: given a source schema S expressed in a source model, and a target model TM, it generates a schema S´ expressed in TM that is "equivalent" to S. A wide family of models is handled by using a metamodel in which models can be succinctly and precisely described. The approach expresses the translation as Datalog rules and exposes the source and target of the translation in a generic relational dictionary. This makes the translation transparent, easy to customize and model-independent. The proposal includes automatic generation of translations.
As a side topic, the same model-independent approach has been applied to other model management operators (merge and diff), leading to the proposal of the MISM (Model Independent Schema Management) platform.
Major papers