Specifying Business Policy Using Agent-Contract Meta-Constructs
Business policy is a set of rules, regulations, invariants and definitions that govern the behavior of an enterprise. It is typically formulated by business people and formalized by lawyers. Business policy is often written down. However, much policy exists only in the memories of those who work for or with an enterprise. Still other policy is embedded in the code of enterprise software systems.
Solution developers have long realized that software is a means of promulgating and enforcing business policy. Without a correct or complete understanding of that policy, software will not function as required by the business. In addition, software that indiscriminately mixes policy and technology is difficult to maintain and reuse. Finally, as the only place where much business policy is documented, software commonly becomes de facto policy.
For all of the importance of business policy to behavior in general and software in particular, it is curious that most businesses have no formal business policy management processes. It is ironic that the haphazard business policy management that is done is almost untouched by automation.
A new meta-model dubbed the “agent-contract” meta-model was created recently by the IBM Consulting Group in collaboration with several of IBM’s industry solution units. This model was created as a means to integrate of IBM’s existing and future industry models. During the development process, it became clear that this meta-model would also provide direct support for business policy analysis and management. This essay describes the major concepts in the meta-model and connects the work to the growing rules/invariants movement within the object community.
KeywordsBusiness Process Semantic Network Commonality Management Business People Business Rule
Unable to display preview. Download preview PDF.
- [B95]Todd Blanchard, Modeling Business Semantics Using Objects and Production Rules, Proceedings Fourth Workshop on Specification of Behavioral Semantics, OOPSLA’ 95.Google Scholar
- [BK83]J. Blank and M.J. Krijger, Editors, Software Engineering: Methods and Techniques. Wiley-Interscience, New York, NY, 1983.Google Scholar
- [D91]Keith Devlin, Logic and Information. Cambridge University Press, Cambridge, UK, 1988.Google Scholar
- [FW87]F. Flores and T. Winograd, Understanding Computers and Cognition. Addison-Wesley, New York, NY, 1987.Google Scholar
- [HKOS96]William Harrison, Haim Kilov, Harold Ossher and Ian Simmonds, From Dynamic Supertypes to Subjects: A Natural Way to Specify and Develop Systems, IBM Systems Journal, Volume 35,Number 2, to appear.Google Scholar
- [M94]Bertrand Meyer, Object-oriented Software Construction. Prentice-Hall International, Cambridge, UK, 1988.Google Scholar
- [MP84]Stephen McMenamin and John Palmer, Essential Systems Analysis. Prentice-Hall, New York, NY, 1984.Google Scholar
- [P95]Marian Petre, Why Looking Isn’t Always Seeing: Readership Skills and Graphical Programming, Communications of the ACM, Volume 38,Number 6, June 1995.Google Scholar
- [R95]Guus Ramackers, Object Business Modeling, Requirements and Approach, Proceedings Fourth Workshop on Specification of Behavioral Semantics, OOPSLA’ 95.Google Scholar
- [S93]A. L. Scherr, A New Approach to Business Processes, IBM Systems Journal, Volume 32,Number 1, 1993.Google Scholar
- [S95]Jean Stanford, Enterprise Modeling with Use Cases, Proceedings Fourth Workshop on Specification of Behavioral Semantics, OOPSLA’ 95.Google Scholar