ERM remains the tool of choice for conceptual data modeling. It relies on entities and relations to model the domain of interest and so does essentially (differences in notation and terminology aside) object oriented modeling. Consistently with this perspective, prevailing “temporal models” are based on facts and their associated valid and transaction times, rather than events and the corresponding occurrence times. Arguably the approach is, in both cases, inadequate for design independent modeling of dynamic (i.e., time-varying) domains and possibly complicates the definition of the system’s behavior.
This paper puts forward an alternative, XCM, that purports to be richer and more domain-oriented, specially where history and forecasting support requirements arising from the temporal nature of the domain of interest are concerned, and that may have a positive influence in other aspects of conceptual modeling.
Bitte loggen Sie sich ein, um Zugang zu diesem Inhalt zu erhalten