IDEA is an ESPRIT project (project number 6333) supported by the European Union which started in June 1992 and was successfully finished in June 1996. The project has been prolongated until March 1997.
This page presents the activities and results of the University of Bonn within the IDEA project. For a complete overview of the project and a list of all partners, you should refer to the IDEA Project Page. Information about the achievements of the IDEA Prolongation Phase can be found at the IDEA WebLab.
The IDEA Team at Bonn is grateful for any comments and helpful discussions on the topics presented herein. It can be reached by...
The goal of the IDEA project was to produce a coherent body of concepts, languages and tools, together with an execution environment, suitable for the design and development of database applications requiring intelligent features. This body of features, referred to as the IDEA technology, has been designed and implemented following a DOOD based approach to achieve the smooth and efficient integration of object-orientation, active rules and deductive rules.
One of the major achievements of the first phase of the project was the design of a common "conceptual interface", called Chimera. The Chimera model is a classical object-oriented data model with operations and inheritance; the signature of objects describes not only the state (attributes) and operations for accessing and manipulating object instances, but also integrity constraints and triggers. The Chimera language integrates operations (used for manipulating objects), deductive rules (used for expressing derived classes, attributes, and views), and active rules (used for expressing database triggers).
|
A Chimera is a monster of Greek mythology
with a lion's head, a goat's body, and a serpent's
tail; each of them representing one of the three language components.
This beast became the mascot of the IDEA project. |
Within the general project framework, research at the University of Bonn has taken place in two main areas, namely the design and implementation of an object-oriented database model incorporating deductive and active rules, and the support of schema design for such a data model, focussing on update propagation, schema evolution, and deductive rule analysis.
Our work in the former task resulted in the specification of the Chimera model and language and the implementation of the Chimera Prototyping Tool which was used as a basis for the development of additional tools supporting the schema design process. Use the following overview to find out more, including the documents describing each of them in detail.
In cooperation with Politecnico di Milano, Italy, we specified the Chimera model and language wich integrates object-oriented, active, and deductive capabilities.
We implemented the Chimera Prototyping Tool (CPT) in order to be able to work with Chimera databases. The CPT is based on an active relational database system. It incorporates an advanced graphical interface which greatly eases the design of complex Chimera schemas.
Access to Chimera databases has been made available to the World-Wide Web with the construction of the CPT/OnLine system. Using CPT/OnLine, you may start a CPT server running at the University of Bonn, Germany, select one of the existing Chimera databases, and then put data manipulation commands for that database. The DML commands are passed over to the CPT server who evaluates them and returns the output, if existing, to the user on the net.
To propagate an update basically means to determine all logical consequences which a physical update of some explicitly stored (base) data induces on any derived data. We studied update propagation for a data model as complex as Chimera and implemented a tool which makes use of Chimera's active rule facilities for automatically performing update propagation.
Schema evolution means the support of schema changes during the lifetime of a Chimera application even if the database is already populated. For the realization of schema evolution, and schema management in general, we strive for a reflective approach which requires the introduction of a meta schema, describing the representation of schema elements as Chimera objects.
The design of deductive rules turns out to be a non-trivial task in the presence of inheritance and redefinition. For example, the concepts of safety and stratifiability have to be extended accordingly. Our Passive Rule Design Tool graphically guides the user through the task of deductive rule design and supports him in the detection and removal of inconsistencies.
The Chimera dialect recognized by the IDEA Tools developed at Bonn slightly differs from the official Chimera specification of the IDEA project. Refer to our Differences Overview for a list of the most important changes.