I've been in the enterprise software business for a long time, and for a long time I've had several related intuitions about how requirements, rules and SOA fit together. But frankly, I never managed to get to a coherent whole about them. Many times while reading this book I kept saying "yes of course, why didn't I think of that?" There are so many excellent insights in this book.
Taylor and Raden may have created a new movement with this work in Enterprise Operational Decision Management. The central theme is that organizations are known by the decisions they make, and not just the major strategic decisions, but the myriad small decisions that their thousands of employees make on a day to day basis. Up until now we had to make due with Decision Support, Knowledge Management, Business Intelligence, Data Warehouses and other off-line aids for manual decision making. In the last few years the maturation of Rules Management systems and the near universal adoption of SOA, Work Flow and BPM are making it possible to more the entire decisioning process into real time, whether human assisted or fully automated.
Two other profound ideas I want to comment on are the champion/ challenger concept, and the role of hypothesis and prediction. Each alone is worth the price of the book.
The champion/challenger concept says once you have a decision model in place and working you owe it to yourself to constantly challenge it by setting up a series of alternate models and running some percent of the decision flow through the challenger model and testing the outcome against the current (champion) scenario. This wasn't really viable until the advent of SOA. They make a great case for how this arrangement allows firms to continually improve their decision making.
A traditional rule system runs off what the experts think the best thing to do in the face of uncertainty. But unless and until a system makes predictions about the outcome of its decisions and closes the loop with the actual results (which of course are often not known for quite some time) it will not be able to improve. This is the heart of their prediction driven decision model.
The book is obviously based on a wealth of information: there must be nearly 100 case study/vignettes sprinkled throughout emphasizing the points just made.
Excellent and inspiring piece of work.