A specification-based data model

3Citations
Citations of this article
1Readers
Mendeley users who have this article in their library.
Get full text

Abstract

This paper presents a conceptual data model for engineered products ranging from software systems to physical objects. The presented model views the hierarchy of components that constitute a system as layers of alternating specification and implementation. If one considers the principles of abstraction and encapsulation, one can see that this viewpoint is quite natural. Abstraction implies that all implementations axe implementations of some specification. Thus specifications may be regarded as directly “above” implementations. Encapsulation implies that implementations, at least conceptually, use specifications rather than other implementations to fulfil] their goals. Thus, implementations may be regarded as directly “above” specifications. This viewpoint has other advantages. It keeps specifications and implementations consistent with one another, models an evolving system nicely, and avoids version percolation problems naturally. It also suggests a way to separate local and global issues in system design.

Cite

CITATION STYLE

APA

Gandhi, M., & Robertson, E. L. (1992). A specification-based data model. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 645 LNCS, pp. 194–209). Springer Verlag. https://doi.org/10.1007/3-540-56023-8_13

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free