An evolutionary approach to application development with object technology

Abstract

Although object-oriented programming (OOP) is not new, it has only recently begun to gain acceptance among independent software vendors. Reasons for this acceptance vary, from a need for basic data encapsulation, to the promise of code reuse, through problem abstraction as a way of dealing with complexity. Despite the advantages inherent in OOP, obstacles to integration with or replacement of existing systems are significant. This is especially true for Application System/400™ (AS/400™) application vendors, because of a tradition begun with the System/38™ of customers demanding source code. Each independent software vendor (ISV) must determine how to make the transition from procedural to OOP languages in a cost-effective way. This must be done for both ISVs and customers, who have often invested heavily in enhancing and modifying source code to meet their business needs. Acacia Technologies has studied the problem and has adopted a strategy aimed at easing the transition by using a phased approach, starting with encapsulation of AS/400 RPG/400™ functions, continued through relocation of modules where appropriate into a multitiered client/server architecture, with a final target of object-oriented modules communicating in a networked environment. This paper will discuss our approach and the part the San Francisco™ project is expected to play in its implementation.