ASPIRE: Advanced Modeling and Specification of Distributed Information Systems

Project duration
February 1997 - March 2000
Funded by
European Union (ESPRIT Framework IV)

Short description
The objectives of the working group are to explore foundations for specifying and developing widely distributed information systems (WDIS) such as those owned by banks, airlines and governments. The explosive growth of the Internet is evidence that a global information infrastructure is developing, and expansion in the use of WDISs can be predicted.
For obvious reliability reasons, WDISs need to be safe and effective. However, their development can be achieved practically and economically only by adapting, extending and (especially) integrating existing software systems. It is becoming standard to upgrade software by `face lifting', in which new interfaces to other systems and perhaps some new features are added to otherwise unchanged software. The working group addresses the need for advanced modelling and specification techniques which are applicable to designing new systems by integrating and re-engineering heterogeneously specified old ones.
The working group will capitalise on the participants' expertise in high-level techniques such as object-oriented design and specification languages and logics. Since information systems are reactive systems, existing specification tools are largely inadequate. The challenge is to cope with inter-system concurrency, synchronization, and communication at an appropriate level of abstraction.
The goal is to reach a level of understanding and agreement among participants as to which methods, language features and background theory should be adopted and adapted, based on experience with high-level specification systems developed within the consortium: OBLOG, TROLL, GNOME, ALBERT, and OBD. It is not the goal to integrate these languages.
Special attention will be given to real-time constraints, deductive capabilities and defaults. Real-time constraints set limits to when an action may or must occur or terminate, and how long it may take from a triggering event to the corresponding reaction. The role of deduction is to predict the effect of a design before it is implemented. Defaults enhance modularity by allowing assertions to be made in a local object, even when the vocabulary needed to specify their exceptions is unavailable.
The approach is to supplement the work done locally and strengthen already existing contacts and cooperations. The academic partners will perform the fundamental research envisaged. The OBLOG industrial partner will play the role of demonstrator by designing the next version of the workbench. The FhG-ISST extra-university applied research institute and the HUP industrial partner will play the roles of (i) users by critically reviewing and commenting the suggestions and results from the viewpoint of usability and effectiveness, and (ii) evaluators by checking the usefulness of ideas and approaches as well as of the design of the future OBLOG workbench. One way to accomplish this is to help in selecting and providing realistic case studies and, if feasible, cooperate in performing these.
The industrial partners will exploit the Working Group's findings for the development and improvements of their products. The potential lies in the development of solid engineering foundations for designing, implementing, reengineering and extending distributed information systems. Progress in this area is prerequisite for improving industrial practice and attain higher productivity in software industry.

Project team
Michael Schrefl (DKE)
Hans-Dieter Erich (TU Braunschweig)