Hans-Dirk Walter

GOM: A Strongly Typed Persistent Object Model With Polymorphism

In this paper the persistent object model GOM is described. GOM is an object-oriented data model that provides the most essential object features in a “lean” and coherent syntactical framework. These features include: object identity, object instantiation, subtyping and inheritance, operation refinement, dynamic (late) binding. One of the main goals in the design of GOM was type safety. In order to achieve this we developed a strongly typed language that enables the verification of type safety at compile time. It is shown in this paper how commonly encountered “traps” for strong typing are avoided in GOM by specifying a very clean subtyping semantics on the basis of substitutability and type signatures.

ECA Functionality in a Distributed Environment

The capabilities of active database systems remain meaningful for distributed applications. However, since the assumptions underlying a distributed active database system differ from those for a centralized active database system, the technical capabilities necessary for supporting active functionality must be revised and extended, and new challenges must be met. In addition, distributed active database systems and distributed rule processing offer new opportunities that make explicit use of the potential of distribution. The goal of this chapter is to provide a systematic overview of the design options for ECA functionality in distributed environments, highlighting the choices from a usage perspective.

Autonomous objects: a natural model for complex applications

Object-oriented database systems are emerging as the “next generation” DBMSs for advanced applications, e.g., VLSI design, mechanical CAD/CAM, software engineering, etc. However, a close analysis indicates that the requirements imposed by these application domains cannot be met by an object-oriented model that relies purely onpassive objects. In this work we go beyond the conventionalsingle-thread-of-control paradigm of passive object models and propose a model ofactive objects which can autonomously initiate responses to environmental changes.Autonomous objects cooperate with each other by synchronous orasynchronous message passing—giving rise to themultiple-thread-of-control in such an environment. It is shown howevents—to which active objects react—can be incorporated into this model. We propose a nondeterministic computational model for the individual active objects that allows the autonomous reaction upon events. We show that this very sparse extension to an object-oriented model gives rise to several high-level features which can be controlled by events. The object-oriented paradigm allows one to isolate the rules according to which events are being raised. This leads to a potentially rather efficient execution model compared to existing relational concepts, which are typically globally-defined event trigger mechanisms.

Activities in Object Bases

Objects are collected into an object base because of a presumed need for cooperation among them. In classical object bases the cooperation is based on synchronous, preplanned message exchange. Many of the modern application scenarios such as industrial and office automation with their high volume of concurrent, interleaved, and iterative actions defy preplanning and require the support of a highly dynamic relationships among the objects.

Structuring the Distributed Object World of CIM 1

Abstract This papers thesis is that today’s object-oriented data models do not offer enough expressiveness to structure the distributed world of CIM in an appropriate manner. Two relationships useful for modeling CIM applications which common object-oriented models lack are identified. The first expresses that some objects describe different prospectives — or facets — of one entity in the real world. The other represents the distribution of activities within a factory. The usefulness of both relationships is argued and a comparison with other relationships which are mentioned in the literature is given.

Integrierter Objektbankentwurf in der Produktmodellierung

Prof. Dr.-Ing. Peter C. Lockemann (2. von rechts, siehe Editorial). Dipl.-Inform. Christoph Kilger (1. von rechts), Dipl.-Inform. Andreas Zachmann und Dipl.-Inform. Hans-Dirk Walter (3. und 4. von rechts) studierten Informatik an der Universität Karlsruhe und sind dort seit 1990 Mitarbeiter am Institut für Programmstrukturen und Datenorganisation (IPD). Ihr Forschungsschwerpunkt liegt in den Bereichen objektorientierte Datenbanksysteme und ingenieurwissenschaftliche Anwendungen.

STRUCTURING THE DISTRIBUTED OBJECT WORLD OF CIM1

Abstract This papers thesis is that todays object-oriented data models do not offer enough expressiveness to structure the distributed world of CIM in an appropriate manner. Two relationships useful for modeling CIM applications which common object-oriented models lack are identified. The first expresses that some objects describe different prospectives — or facets — of one entity in the real world. The other represents the distribution of activities within a factory. The usefulness of both relationships is argued and a comparison with other relationships which are mentioned in the literature is given.