Reuse of Objects in Models Being able to reuse objects directly by reference is
an important part of the productivity gains of DSM. The same object can be visible in
many places, but in each case it really is one and the same object. An object can be
reused in many different ways:
. As a property value of several objects, as seen above.
. By appearing several times in the same graph representation, as a notational
convenience. In a diagram, this can be to prevent lines crossing or becoming too
long, as in the Watch examples in Chapter 9. In a matrix, the same object will
often be found on both axes, to allow relationships with itself to be shown.
. By appearing once in each of several different representations of the same
graph. This allows radically different views on the same graph, for example, as
both a diagram and a matrix, or as one diagram emphasizing inheritance and a
second diagram emphasizing aggregation.
. As an element of several different graphs of the same modeling language. This
allows a similar kind of reuse to functions and components in textual programming
languages. See, for example, the states representing WatchApplications
in Chapter 9.
. As an element of several different graphs of different modeling languages. This
allows method integration: building a large model with different viewpoints or
aspects described in different modeling languages.
Pages:
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708