Methodical Resolution of Pivotal Issues

Addressing a timely and vital topic, 'Modernizing Legacy Systems' is an excellent book from the standpoints of both content and presentation. The advocated approach, which is termed the Risk-Managed Modernization (RMM) Approach, is synopsized on page 28 in UML activity diagram form. The exposition in the subsequent chapters is keyed to corresponding activities in this diagram. This key makes it quite easy to situate and interrelate the coverage of the respective chapters in the context of the RMM Approach. I found this recurring orientation feature to be very helpful in understanding and integrating the book's content.

The book describes, rationalizes, and selectively illustrates the RMM Approach, where the continued availability of the legacy system capabilities is necessary over the sequence of modernization increments. While the approach is illustrated through an incremental transformation of a legacy COBOL-based system to a Java-based derivative, the RMM Approach is nevertheless applicable to other modernization problems or technologies. Moreover, the book does an exceptionally good job of interweaving explanations with examples. These examples are modest but salient and revealing, thereby avoiding unwarranted detail or distractions.

The advocated approach is at once both architecture-centric and component-centric. Architecture centricity captures and sustains a rather specific vision of the as-desired system, and the associated target architecture provides a stable reference over the various modernization activities. Component centricity enables the identification, analysis, grouping, and ultimate realization of system elements that are allocated to the respective modernization increments. Overall then, the target architecture establishes the initial and termination points of a modernization project, and the componentization installments determine the actual redevelopment trajectory connecting the project end points.

For me, the most intriguing, innovative, and vital parts of the approach appear under the RMM activities labeled Define Modernization Strategy and Reconcile Strategy with Stakeholder Needs (Chapters 13-15 and Chapter 16, respectively). Basically, the modernization strategy provides a systematic approach to delineating, analyzing, and grouping modernization elements through an examination of the legacy system implementation, subject to project constraints and certain prior higher-level technical decisions. Then, the finalization of element groupings into sequential increments is determined using programmatic preferences of the various stakeholders. This two-stage definition of modernization increments is driven prominently by cost and risk considerations, as well as by programmatic and technical factors. Ultimately, the designated increments establish waypoints on the aforementioned redevelopment trajectory, thereby identifying interim architectural configurations that facilitate closure on the target architecture, while simultaneously maintaining user capabilities during the modernization effort.

In all, 'Modernizing Legacy Systems' is a readable, coherent, illuminating, and surprisingly broad treatment of a vital topic. Hopefully, the RMM Approach or variants thereof will see widespread use in industry, thereby exploiting "a systematic and fact-based method that avoids arbitrary, intuitive decision making..."

Other Reviews

• A Legacy Migration Classic
• Methodical Resolution of Pivotal Issues