/pdf/the-morgan-kaufmann-series-in-data-management-systems-47lgtms9oc.pdf

The Morgan Kaufmann Series in Data Management Systems (Selected Titles)

Over the years, IBM has made progress in resolving many of the issues that deal with improving application development (AD) productivity and quality. Systems Application Architecture™, together with IBM's recently announced AD/Cycle™ direction, provides a platform for even greater progress. This paper addresses the IBM strategy that supports AD/Cycle and gives an overview of the major components of the AD/Cycle architecture. This paper is an introduction to other papers that follow in this issue.

AD/Cycle strategy and architecture

Critical success factors in data management

This article describes a three-month pilot project to model and re-engineer an internal business unit of a major multinational firm using Enterprise Analyzer, a methodology and supporting group software tool for modeling and analyzing business processes. The methodology combines collaborative model building with creative problem reframing to enable multilevel, cross-functional business teams to quickly examine and redesign business processes. The tools developed for this project extend the capabilities of existing electronic meeting system technology to help groups build complex representations of their organization. The results of this pilot project suggest that the methodology and supporting tools can be effectively applied to re-engineer business processes, but also suggest the need for better ways to help groups accept innovative ideas. Many innovative ideas were generated, but few made it past the participants' evaluation and into the project's final recommendations.

Methodology-driven use of automated support in business process re-engineering

Quality is probably one of the most serious concerns of today's software community. For software applications exhibiting a certain complexity, the quality of a product can only be guaranteed by a methodological approach, using appropriate administration and tools. The methodology and administration must be manifested in a well-defined and well-observed application development process. The process must integrate the human activity, the tools, and the intermediate and final work products into a coherent flow of actions. In this regard, the development of applications follows patterns that are well established in other industries where an application development (AD) process model is defined and then executed via an interpretation mechanism. The complexity of the development process makes it necessary to support and integrate all of its aspects by means of on-line interactive computer support. Computer-aided process support in the general sense we call work management. This paper explains the concepts of an application development process model and of work management for application development under AD/Cycle™ and its relation to project management.

The role of work management in application development

One of the motivations for the use of a facility such as the Repository Manager™ in an information processing system is to centralize the information needed for the development of software. What this information is and how it is interrelated is defined in the underlying data model. This paper discusses the kinds of information required for software development and offers some suggestions on how the data model should be organized and implemented.

R. W. Matthews

Papers

Data modeling for software development