What’s the Worst that Could Happen? More Bad Assumptions, Ignorance, Failures, and Screw-ups in Engineering Projects. Volume-II (Computer Architecture)

Unfortunately, we learn more from failures than from successes. This book continues an examination of a cross-section of engineering failures, and analyzes them to define the lessons-learned. It also presents some additional methodologies to prevent failures, or, at least, minimize the effects. The first volume presented a cross-section of failure studies, mostly drawn from the engineering and aerospace context. Each study includes specific references and a definition of the root cause of the failure. I said, "Let's try to learn from other's mistakes. It is less painful to learn from others' failures than your own." So, almost nobody listened, and I have enough material for a new book. We will see continuing errors of omission, errors of commission, and just plain ignorance of the facts. Since its publication, I have had more than enough material for a second book. The first book discussed System Engineering processes and procedures that can and should be applied to an architecture before the fact, and, unfortunately, after the fact as a post-mortem analysis. It is important to do a good post-mortem analysis of failures, and document them, for the benefit of the next generation of implementers. This helps to prevent the repeating of mistakes. And yet, we keep repeating mistakes, not building on our work, and going on to create larger, more creative mistakes. I will try to minimize the overlap in the books, but some times there is further information or corrections on material in Volume 1. A major new area of interest is self-driving cars, not just their technology, but the insurance and legal issues they introduce. I am presenting this from an engineering standpoint, because that is my background. But the concepts apply across all disciplines and endeavors. If your are responsible for a design, a device, a policy, or a program, you must think through the consequences. Always have a plan B. Always have a Plan C Always think about safety and security. Don't make the same mistake twice. Don't make a second mistake. I have included an updated glossary and a list of reference material at the end of the book, and specific references for the failure cases discussed.