To Forgive Design: Understanding Failure Hardcover – March 30, 2012

Petrosky (professor of civil engineering and history, Duke University) has produced another winner in this handsomely produced book from Belknap/Harvard. (The book jacket is particularly handsome, as well as to the point.) Petrosky is that rare bird, an engineer who writes about his subject in a way that not only practitioners but lay people can appreciate. He's not a great stylist but his prose is lucid and he always says something worth saying.Across several books, the author has pursued the same two central preoccupations: (1) the history of engineering and industrial design and (2) failure analysis and its relevance to the advancement of the discipline. On history, The Pencil (1990) is particularly fascinating, but so are his essays on the evolution of the pencil, paper clip and silverware in The Evolution of Useful Things (1992). Engineers of Dreams (1995) celebrates American bridge builders. A later book, The Toothpick (2007), uses a humble household implement to explore the complicated interaction between societal demands and engineering design challenges in the creation of a new or improved product.This present book revisits the preoccupations of his 1985 book, To Engineer Is Human: The Role of Failure in Successful Design. But there's enough new in it -plus Petrosky has seventeen more years of experience to draw upon--to make it worthwhile, even for someone like me (not an engineer) who has bought many though not all of Petrosky's books and read them avidly. Essentially, Petrosky's message is that engineering is a self-correcting discipline but as new solutions arise, engineers grow complacent as a result of success. They start pushing the boundaries -lowering the safety factor, shaving costs, cutting back on inspection--and eventually create a new failure . . . from which subsequent students learn a lesson and make corrections, and so on and so on, ad infinitum.Petrosky has repeatedly demonstrated how much he likes bridges, but bridges are a useful kind of construction to make the points he makes about how his discipline advances out of analyzing its excesses. Failure is more than useful in engineering: it creates the conditions for new advances and new enlightenment, new."Galloping Gertie" -the Tacoma Narrows bridge that shook itself to pieces shortly after it opened for traffic in November, 1940--is an example. With the completion of the George Washington Bridge in 1931, the esthetics of bridge building morphed across the country: everyone wanted to build bridges that were slim, graceful and extended. The Tacoma Straits bridge was an unanticipated consequence of that trend. With only two traffic lanes, the new bridge was too narrow to dampen cross-wide vibrations engendered in it by winds and the steady flow of traffic across it. Even before the bridge opened it had acquired its nickname. On particularly windy days, commuters waiting in line to cross the bridge could see cars ahead of them rise and then dip with the bridge's undulations. Then came a 42-hmile-an-hour wind. The bridge's central span collapsed, all 2,800 feet of it. No one was killed, fortunately, but they could have been. Out of that failure came a deepened appreciation for, and soon more sophisticated understanding, of the impact that exaggerated aerodynamic forces can have on a rigid/flexible structure like a bridge.Petrosky discusses other bridge failures -from the Tay bridge collapse in Dundee, Scotland, in 1879 to the collapse of the Silver Bridge, connecting Point Pleasant, West Virginia, and Gallipolis, Ohio, in 1967. A lot of the book is about bridges but not all of it. He comments on other kinds of failure as well: the collapse of the pedestrian walkway in the Hyatt Regency hotel in Kansas City, Missouri, in 1981; the disintegration in mid-flight of the space shuttle Challenger in 1986; the humungous oil spill caused by the explosion on the oil-drilling rig Deepwater Horizon in 2010. There's a chapter on construction cranes. On an average, he writes, almost ninety people die yearly as a result of "crane accidents and related hazards." (When I lived in Dubai, from 2001-4, I used to joke that "if there is a national bird for the Emirates, it's the construction crane.") He has insightful comments as well on the difficulty of predicting load strain on pedestrian footbridges.Petrosky urges his confreres to recognize the importance of past history, and past failures, in learning and practicing their field. "The surest way for the designer of any system to achieve success is to recognize and correct the flaws of predecessor systems, whether they be in building codes or in banking practices or in bridges."In engineering, as in Greek tragedy, hubris can prove fatal.

Laborious. Tedious. Melodramatic in spots. Dis-ordered. Much irrelevant wandering. Certainly I was disappointed with this book. I bought it on account of a positive review in the Wall Street Journal. This is the 2nd WSJ review that has led me astray - I can't recommend those either.Perhaps 150 of the 360 content pages would have been eliminated by a competent editor. An entire chapter is devoted to mostly idle chatter about the author's graduate studies and experiences in a lab "known affectionately by the acronym TAM", a vignette of his dissertation advisor, and recountings of coffee klatches. I am an engineer by training, trade, and practice of more than 20 years, with plenty of time spent in labs. I did not develop an attachment to any of them. Another chapter is devoted to a meandering history of "Iron Ring Ceremonies". The first page or so is interesting - I had never heard of them - but little is gained thereafter. And then throughout the text we find such gems as "Success is success, but that is all that it is." Where was the editor?I was hoping for an organized synopsis of failures of various kinds with details of the believed causes, as well as concise discussions of the non-technical human factors that are involved in almost all of them. It's not here. The narrative is often of the jumbled stream-of-consciousness type, with the author dropping into first person and diverging into all manner of side-topics. We have tortured discussions about bridges, ad infinitum, with revelations such as "Among the most important decisions in designing a new bridge are where to locate it and what kind of bridge to build". Luckily I bought this book. In some cases (e.g. the Columbia break-up and the Hyatt Regency walkways), the important details of the failure mechanism are not even explained. There are no sketches, no drawings, no tables of comparative data, no statistical summaries, and only a few photos - several being portraits of professors and colleagues in lieu of failed components. I have read technical reports on mechanical failures that were better organized and more thorough - and much shorter.Credit the author with adequate referencing - there are copious footnotes. And there is some good, and to me new, information to be gleaned from this book, but you will need a fairly wide sieve and a tight dust mask - the volume of chaff can be overwhelming.

This book provides lucid explanations of several spectacular failures from the Titanic to the Tacoma Narrows bridge and places them in an intellectual context that should be helpful to many a young or future scientist or engineer. It is written as if an avuncular uncle of great knowledge and experience were talking to a group of students. It is not tightly edited and formatted in Power Point logic. This may distress some purists, but they will miss an enjoyable and potentially valuable few hours in their obsessing over the paragraphs or even chapters they find distracting. My advice is to read the book for the 80% which is difficult to find elsewhere and enjoy the autobiographical excursions for what they are.The truth is, engineers and scientists have few general audience voices for their craft besides a few writers at places such as The New York Times and The New Yorker. No one writes much about civil engineering and no one writes about learning from failure as a cautionary tale, not against progress, but against the arrogance which can stop progress for years if not decades. John McPhee and the two generations of his students simply can't cover all subjects.

As an engineer, I was looking for 'a philosophy of risk and failure in engineering' which might carry over to complex software/electronic system design. This book is more a beautifully written long and detailed description of notable engineering failures. There are useful insights into failure ( for example, understanding that understanding failure is an essential aspect of a successful engineer). But you do have to wade through some pretty lengthy tangantial material.At its worst, its a ramble about his working environment and colleagues. At its best, its an insight as to how often it is management which fails, whilst concerns of the engineers in the ranks are pushed aside for managerial 'goals', thus leading to an 'engineering' failure. (e.g. The Challenger disaster).Despite its flaws, it makes it into my library for the sometimes even warm and enveloping writing (yes!) and the occasional useful insight, even after lots of reading.If you are interested in risk more generally, and rather from the top down rather than bottom up, have a look at Risk by John Adams:  RiskRisk

I understand this is a follow on to an earlier book. I think he needed to top his pension up or prove he "still had another book in him".I'm interested in the subject and I found it a rambling, ill organised collection of war stories. I think he just published his notes. Needed a much more aggressive editor.Some gems and points of genuine interest scattered throughout the book.[needed a stronger editor? I find myself saying this a lot these days...]

This is a reasonable introduction to the issues in robust design, but it is neither well-written nor well-edited. It's rather repetitive, and many of the examples are very familiar - the O rings in Challenger, the Tacoma Narrows bridge, various aircraft disasters - so the lessons aren't fresh. There is a good book to be written on this topic, but it's shorter than To Forgive Design, and more interesting.

Some very interesting stories, but little too much "water"

Bit of a slow read, not finished it yet