Foreword

You’ve probably heard about the South Sea Island cargo cults, but let me refresh your memory a bit. Some good examples happened during World War II when military people set up supply chain airbases on some of the islands. The native people watched the military people set up radars and radio telephone boxes and phones. Then the military people would talk into the phones and airplanes would fly in with cargo, which was shared with the native people.

When World War II ended, the military people packed up their equipment and left and no more airplanes or cargo came. So the native people attempted to bring the airplanes and cargo back by setting up wooden radars, boxes, and phones, dressing up in uniforms, and shouting into the phones. But no cargo came, and they were mystified and frustrated.

Unfortunately, the same cargo cult syndrome can and often does happen with Capability Maturity Models (CMMs), including the CMMI^�. CMMs capture processes used by successful organizations and surround them with such modern cult appurtenances as trademarks, service marks, and licensed assessors. Unsuccessful organizations that want to develop on-time, within-budget, high-quality software and systems often set up these processes and expect that by following rigorous procedures for requirements management, planning and control, and quality assurance, that on-time, within-budget, high-quality systems will emerge for them too. But all too often, very little useful “cargo” appears and the organizations are mystified and frustrated.

In this book, Michael West does a great service for many potential CMM or CMMI adopters by explaining the differences between pro forma and in-depth organizational process improvement. He also illustrates the differences through many examples from hard-won personal experience as a process group member and process improvement consultant for a wide variety of organizations. He shows the difference between attacking symptoms (e.g., trying to use requirements management change control procedures to eliminate scope creep) and addressing root causes (e.g., poor customer communication, overinterpreting “the customer is always right,” lack of change impact analysis). He debunks various myths about CMMs (e.g., higher maturity levels will fix your problems and guarantee success, organizations operating without CMMs are in total chaos, CMMs are expert-only “rocket science”), and provides case studies from experience showing that the myths are dangerous to believe.

More than this though, he provides positive guidance on how organizations can draw on the CMMI or other CMMs to effect real process improvement. He emphasizes that most organizations have some highly effective practices that fit their culture and business environment and shows that these need to be understood, conserved, and built upon rather than being trashed in a “slash and burn” approach to process improvement.

He provides a good project-oriented sequence of steps for addressing the root causes rather than the symptoms: establishing a common language, determining your organization’s starting point, establishing goals and success criteria, planning the improvement project, and monitoring and controlling its progress.

Each chapter has some good features for keeping the reader oriented and continuously engaged, starting with a “What Do You Think? What Do You Believe?” assessment, summarizing what to do and what not to do, and ending with a “What Did You Learn? What Will You Do?” assessment.

If you’re contemplating any sort of process improvement effort, this book can help you a great deal in defining and getting the results you’d like to achieve.

Dr. Barry Boehm

TRW Professor of Software Engineering,


Computer Science Department


Director, USC Center for Software Engineering