Status and conclusions

In this paper we have discussed dynamic models in relation to static models, as they are used during an object-oriented design process. It is a central claim of the paper that the dynamic model should be constructed before the static model, especially in complicated design situations. Whether a design situation is complicated depends on the skills of the designer. The main purpose of our research is to find out to which degree the claim is true.

A design model can be characterized with respect to

the means of expressions
the degree of formalization, and
the level of abstraction.

Our proposal for a dynamic OOD model avoids a fixed stand on the first characteristics, because we chose an abstract, internal representation which can be presented in many different ways, using several different means of expressions. We go for a dynamic model with both formal and informal aspects. In several respects, the informal and formal aspects describe the same elements. Finally, the level of abstraction is modest in our proposal. The starting point of the model was the actual program execution model, on which we abstracted irrelevant details away. Consequently, our dynamic model has similarities with an actual execution of an object-oriented program.

We do not think it is possible to make satisfactory descriptions of dynamic OOD models as diagrams on sheets of paper. Therefore we have proposed to deal with dynamic models on a more dynamic medium. The medium we have described is a dynamic exploration tool. With the tool we have in mind, it is possible to build, explore, and analyze a dynamic model.

Currently we have implemented a very simple prototype of a dynamic exploration tool in CLOS. We are also in the process of implementing a graphical user interface to the tool. We plan to make an additional and improved prototype in order to gain more experience, and in order to test the hypothesis, which have been formulated in this paper.

Next: References Up: Dynamic Models in Object-oriented Previous: The work by

Kurt Noermark
Mon Feb 26 11:36:35 MET 1996