Preface Preface Almost two decades ago, I completed a project to develop and deploy a teller and sales application for a large U.S. bank. Enhanced business capabilities, technology upgrades in the branches, and a pending bank merger were the business drivers. Some months after the production roll-out, as the Chief Architect, I was invited to a meeting with the Vice Chairman of the Retail Banking who wanted to understand my perspective on how the bank should address challenges in meeting future demands that required extending the reach of the teller and sales platform functionality to other parts of the bank. The Vice Chairman was responsible for all retail functions of the bank and expansion was hot on the agenda. The bank was growing, entering new market places, acquiring banks, opening branches, and rapidly attracting new customers. We sat down and discussed cross selling, expansion goals, and the need for several parts of the Bank such as credit card processing, wholesale banking, and loans, to be able to access and use functionality contained in the teller application we had just built and deployed.

Obtaining customer information, account balance inquiries, and address updates were just a few of the basic pieces of functionality needed by these other departments but there were more complex pieces of business functionality required, too. When wholesale banking or credit card processing needed to access data or functionality in the teller system, they needed to go through a development cycle that necessitated waiting in a queue with others, whereby the IT department could prioritize and satisfy the multiple requests and requirements. The Vice Chairman expressed this current situation as a problem; it impacted the bank's capability to get more products out the door faster and his ability to meet sales and revenue targets. He asked two questions: How can we do this better and how can the bank provide access to previously built and deployed business functionalities to other parts of the bank without going through IT development queues? Addressing this question and others by senior business executives has been top of mind for me for two decades. Over the last decade, I have met with corporate executives from hundreds of companies across the world whose enterprises are characterized by disparate and siloed systems and applications; horizontal integration is the goal as businesses seek greater agility in the global marketplace. Corporate managers are asking how do to make the IT system more flexible so that it is easy to connect across the enterprise and so it is inexpensive in both time and cost. The story of the bank occurred two decades ago, but I find CEOs and other corporate executives asking this same question over and over again, decades later. Everyone is searching for flexibility as competition intensifies.

Everyone sees this albatross around their neck getting uglier and negatively impacting goals for growth and limiting the responsiveness and agility required as the cost of maintaining, integrating, and supporting systems is rising. Less capital is available for innovation, changing the business, and delivering new capabilities. Just a few years ago, I met with a corporate manager responsible for a payments business. His frustration was apparent as we discussed the need to change his three-year-old IT system to accommodate new channels (phones, kiosks, and other mobile devices) and new market segmentations. He was frustrated because although he was not a technologist or software engineer he knew something was not right. He wanted to know why after millions of dollars of investment in a creating a new payment system, built three years earlier, it was not easy to change his payment system to accommodate small and medium businesses or to allow access to payments using handheld devices. He asked this question because his payment system was built with modern software engineering best practices yet flexibility was evasive: adding new channels and new customer segments would take too long and cost too much money as if he were building the system from scratch versus just changing the system. I responded and the short answer is that applying best practices and modern system engineering practices is not sufficient if agility is the goal.

I further stated that there is a considerable amount of data that shows this problem is not isolated that most applications become difficult to change within 3 to 5 years after the first production deployment. Recently, I was in Mexico City working with a large logistics company. It was just finishing an 18-month project to reengineer a core IT system that was no longer responsive to the business. The new system was engineered like the bank system two decades earlier, with the best software engineering practices and tools available. I was asked if this new system would suffer the fate of past systems in its capability to be responsive. That is, would this system become brittle in the future and if so, why? Would this new system be built for change such that flexibility was an attribute of the system and not a platitude? Again I answered no, stating that applying best practices alone will not achieve the goal of agility. I know this is true because his team and teams just like his around the world have been using modern and best practices of software engineering for years with the same results. The result is that three to five years after the system has been deployed it is difficult to change, and is expensive in time and money.

It is not only the commercial world that sees a problem but the public sector. we have met with various public sector organizations over the years and my interactions confirm that they are confronted with the same challenges we see in the private sector. :Public and private sector managers see the rising cost of support, integration, and maintenance of the systems as a ball and chain that is a huge drag on cost reduction and as a result, it puts a limit on monies available for creating new capabilities in the theater as the available dollars are limited. It is these questions and their answers that prompted us to write this book about service-oriented architecture (SOA). This is not a technology book, but a book for technologists and business stakeholders. We hope to demonstrate, that SOA and service-orientation in general, is not solely a technology play but a paradigm and architecture that calls for business and IT collaboration and when understood and applied, it can change the course of your business, where flexibility and lower total cost of ownership become realities. Total cost of ownership and flexibility are different sides of the same coin. There is less flexibility when funds are not available to spend or when providing new capabilities is constrained because resources are consumed in integration, maintenance, and support.

Flexibility is evident when the business, not IT, has the power to deploy new business features without IT development queues or when new capabilities can be provided in weeks or months instead of years, and when two or more capabilities can be composed at will to create a new, enhanced capability that directly supports business drivers and alleviates painpoints. If we make the right choices, we will have a chance to escape from the boxes that frustrate us today. The escape will not be easy -we will be constantly challenged to question conventional assumptions and comfortable practices. Many will not even see the opportunity. They will continue to remain closed in the boxes that make every day more frustrating. Some will see the opportunity but will either try to move too quickly or fail to stay the course. They will blame the technology for its failure to produce results. For those few who succeed, the rewards will make the journey well worth the effort.

-John Hagel III in Out of the Box Our choice as managers, leaders, or architects is to seize the opportunity and release ourselves from self-imposed boxed thinking because "if you don't change anything, nothing changes." We can make business flexibility a reality with IT support but it requires a vision, a strategy, execution of the strategy, and most importantly, staying the course. The strategy must be a living plan accompanied with a evolving roadmap that can be implemented, monitored, and measured. It requires you take incremental steps that together bring about change: incremental and quantum leaps over time. If you want to get out of the boxes that John describes, enabling your IT systems to be engines of innovation, this book will be of value. If you are responsible for strategy in the organization and need to link that strategy to an IT strategy to make your IT systems and infrastructure capable of supporting a rapidly changing landscape or business model, you should read this book. If you are tired of reading about platitudes and seek guidance about how to achieve business flexibility through the adoption of SOA, you will obtain value in reading this book. This book is not about a technology chan≥ it's about a business journey with IT, where SOA is both the enabler and the catalyst.

This book is different than other books on SOA as content is organized into 100 questions and answers. Feel free to go directly to the chapter that most interests you or go directly to a question for which you would like an answer. Visit http://www.100Questions.info and submit any questions that remain unanswered. Kerrie Holley © Copyright Pearson Education. All rights reserved.