To Engineer Is Human: The Role of Failure in Successful Design

Here it is. I purchased this for my Dad, 79 years young. Finding anything for him, that he hasn't seen, heard or done is very difficult. Imagine my surprise when he sent me a note stating this was the BEST gift he'd received in a long time! I also bought him "the Evolution of Useful Things"... same message! Both are super-duper gifts!

The answer is well explained in this book. By the time you built such a car, it might be so heavy it couldn't move!

The real interest in this book are the analyses of various disasters that should have been planned for, but weren't.

The most terrible engineering disaster (and the reason I bought this book) was the collapse of the sky walkway in a hotel in Kansas City in the 1980's. I was just returning from KC when I heard the horrific news on the radio. The skyway collapsed during a dance, killing hundreds and injuring more in a dreadful disaster. I was very upset by this terrible event. Why did this happen?

The explanation in "To Engineer Is Human" is really brilliant; the walkway was designed "properly" with a bolt that went through the beam supporting it. But it could not be built as designed because the bolt couldn't be installed in the vertical support. Instead, the builders split the vertical support into two parts in order to install two bolts, and each part was then able to move independently, causing a shear force that eventually led to the disaster. A brilliant analysis and one that showed that despite correct design, the plan must be able to be implemented to work--or else the execution of the plan may doomed to disastrous failure.

That lesson is really important when you are engineering anything, even software. You may specify an important feature, but if the R&D department cannot implement the plan, the product may fail to meet its goals, even be defective.

The book is a bit "thin"--I wanted more and wished it were longer and had more detail, but I will say it makes its point and memorably so. After reading it, your eyes will be opened to how things are designed, how things fail and how engineering affects our lives.

This book gives insight into how plans fail. If you are involved in any field in which plans are made and then implemented, be they in any of the engineering disciplines or in any field of human activity, then there is something to be learned from this book.

But the writing is such that you will have to expend energy to extract the message from the text. The author apparently wanted to show off his vocabulary and 'erudition', so it is not an easy read.

Educational but not suited for light or relaxed reading.

The author took 232 pages to state what could have been said in 23 pages. He uses nursery rhymes and Greek mythology to "explain" the publics expectation that some machines and structures fail!?! He would have been better served going into more details of the few cases he revisits repeatedly. As a working engineer, I found his explanations of how complex systems should be developed very off base. What happened to incremental design improvements, testing, instrumentation, quality control, and design reviews?

To Engineer is Human is a surprisingly relevant book, despite being 15 years old now. Some of the examples may tax the memories of younger engineers and engineering students, but that's exactly the point of this book, to emphasize the nature of engineering: improving what has already been done in the past.

I, too, found the repetitive references to a limited number of examples tiring; I suspect this was done because Petroski had prior knowledge of these case studies and wished to minimize his research by drawing on what he knew about before writing. As an amateur historian of technology, I was also disappointed that few earlier historical examples were treated in any depth, the Crystal Palace being a notable exception.

The book is an easy read. Henry Petroski's prose is easy to grasp and flows well, holding the reader's interest, despite the repetition.