Fatal Defect: Chasing Killer Computer Bugs [Paperback]

Ivars Peterson (Author)

Book Description

Publication Date: April 30, 1996

An airplane crashes, killing eighty-seven passengers. A cancer patient receives a fatal dose of radiation from a machine designed to be foolproof. The ATMs at a New York bank debit customers twice their actual withdrawals, resulting in a loss of millions of dollars. In every case, the culprit was a computer bug, a software error or design defect that may escape detection until it erupts into the real world with sometimes catastrophic results.

This arresting and at times terrifying book tells us just how prevalent these defects are and how they are multiplying as computers become more sophisticated and more deeply embedded in our daily lives. It is also a riveting portrait of the men and women who find and "exterminate" those bugs, whether they occur in pocket calculators or nuclear reactors.

Fatal Defect reveals what you should know about the computers in our lives. Read it before you buy a computer, use a cash machine, or book an airplane flight. Then pray that one of its real-life heroes was on the job.

Editorial Reviews

Amazon.com Review

In Fatal Defects: Chasing Killer Computer Bugs, Ivars Peterson describes dozens and dozens of hoary computer bugs and gives biographical sketches of the bug detectives who located and fixed them. This book, which reads like a novel, is both entertaining and informative. Many of the bugs that Peterson discusses are not in computer programs per se but in the human systems that run and operate the computers. Very often the operator fails to understand what the computer program requires as input and types in an incorrect command. The computer then executes the command, with potentially disastrous results. Fatal Defects has important lessons for both those who design computers and those who use them.

From Booklist

When the computer crashes, taking the plane, spacecraft, or phone system down with it, the postmortem falls to "bug hunters," software experts whom Science News reporter Peterson has followed through some of their investigations and analysis. But nonhackers needn't expect an overload of cyberspeak: Peterson explains in clear English how the engineers find and squash bugs. The methods vary with the age of the beast: "program archaeology" involves reconstructing what the original programmer was thinking of, whereas new programs, especially those written for aerospace purposes, utilize "program verifiers" that monitor the program's status. But aside from methods, software-caused disasters are the eye-catching anecdotes Peterson offers, ranging from a radiation machine that killed patients, to France's accident-prone Airbus plane, to Russia's embarrassing loss of two Mars probes, to a disaster-in-waiting--come the year 2000, warns one bug hunter, "algorithmic anarchy" might reign, for business software wasn't designed to switch dates from 1999 to 2000. However shortsighted that design decision was, it well illustrates Peterson's theme: software defects are inevitable. Gilbert Taylor --This text refers to an out of print or unavailable edition of this title.

See all Editorial Reviews

Product Details

• Paperback: 288 pages
• Publisher: Vintage (April 30, 1996)
• Language: English
• ISBN-10: 0679740279
• ISBN-13: 978-0679740278
• Product Dimensions: 5.5 x 0.6 x 8.5 inches
• Shipping Weight: 13.8 ounces (View shipping rates and policies)
• Average Customer Review: 4.8 out of 5 stars  See all reviews (4 customer reviews)
• Amazon Best Sellers Rank: #1,328,933 in Books (See Top 100 in Books)

More About the Author

Discover books, learn about writers, read author blogs, and more.

Customer Reviews

Most Helpful Customer Reviews

15 of 15 people found the following review helpful:

5.0 out of 5 stars Learn from software failures, December 31, 1999

By 

Bret Pettichord (Austin, Texas) - See all my reviews
(REAL NAME)   

This review is from: Fatal Defect: Chasing Killer Computer Bugs (Paperback)

Henry Petroski has written several books that explore his dictum, "Form follows failure." His thesis is that improvements in engineering are made to overcome the failures of previous design. Petroski's books cover advances in civil and mechanical engineering. Ivars Petersen has written a similar book covering some of the notorious failures in software engineering and the efforts by a few leading engineers to define practices and design methods that can prevent such failures from recurring.

Fatal Defect describes dozens of software failures, how they happened, and the efforts to correct them. The defects occur in banking systems, stock exchange mechanisms, aircraft and spacecraft guidance computers, medical equipment, telecommuncations, and scientific computation. Some of these failures are famous; others are little known. Regardless, the descriptions always provide the kind of technical detail that you need to really appreciate the situation. Petersen is a journalist for science news and is clearly a professional when it comes to describing technical issues for the intelligent layman.

Moreover, he tells the stories of people who found the errors, lead the efforts to correct them, or who tried to raise the standards of the industry. Nancy Leveson investigated the Therac-25 defect that lead to several deaths in 1986. This influenced her efforts to design software safety standards. Learning from failures requires knowing about them. But the details of many failures are often kept quiet, being marked proprietary or secret to avoid embarrassment or litigation. Peter Neumann tried to open up the discussion of computer failures with RISKS digest. He started it in 1985 but even today it remains one the best places to learn about the technical details behind dangerous system failures. David Parnas took the lead in criticizing the Star Wars strategic defense initiative. He noted that there would inevitably be defects in the software and that there was no way to conduct a comprehensive system test, short of a nuclear war. He then moved on to overseeing the engineering processes at the Darlington nuclear plant, ensuring that the software was correct, even though this delayed the project completion by three years. Vic Basili was one of the first to conduct controlled experiments with programming teams to determine which development methods actually produced the most reliable software. For example, in 1982, he established that code reviews were far more effective than functional testing, a result that is only beginning to be regularly applied to engineering practice today. These results lead he and Harlan Mills to develop the Cleanroom process which Mills taught at NASA and IBM.

Petersen tells the stories of these and other engineers, describing their background and how their careers lead them on the paths that they ended up. I'd been familiar with the ideas of many of these people, but i found it very interesting to learn of the experiences that had lead them to formulate and articulate these ideas.

The issue of what constitutes effective means for developing reliable software is becoming more than just a practical matter with recent events. Earlier this year the Texas board of professional engineers started licensing software engineers and the Institute for Electrical and Electronics Engineers plans to start certifying software engineers in 2000. Licensing means more than just professionalism and status. It also means acknowledging accepted practice and deviating from it at the risk of malpractice. If the licensing process is done well, it will base itself on the fine, but tentative work done by the people described in this book. If it is done poorly, it may merely enshrine the latest fad in law.

3 of 3 people found the following review helpful:

5.0 out of 5 stars Thought-provoking, November 13, 2002

By 

Kevin B. Cohen (Denver, CO) - See all my reviews
(REAL NAME)   

This review is from: Fatal Defect: Chasing Killer Computer Bugs (Paperback)

From the subtitle, "Chasing Killer Computer Bugs," you would think that this would be a book about software testing. It isn't. Rather, it's a book about things going wrong with software. The author tells readable stories about some infamous software bugs, at least one of which I've never been able to forget. This is the story of the Therac-25 radiation therapy machine, which caused deaths and injuries due to a software problem. Some of these stories have morals which software professionals would do well to keep in mind--and not just developers, but the people who give them their marching orders. For instance, the Therac-25 story makes clear in a very sobering way how an apparently VERY minor change to a program can have VERY unanticipated consequences. The A320 story makes a convincing case for thinking hard about user interface design.

Other parts of the book talk about why building good software can be so hard, and about some of the people and organizations that work towards developing approaches to issues in software quality and construction. You wouldn't think that these would be particularly interesting subjects, but for the most part the author makes them come alive.

This is not a technical book--don't expect to come away from it with any new debugging techniques. Rather, expect it to give you lots of food for thought.

2 of 4 people found the following review helpful:

5.0 out of 5 stars Well thought-out, October 2, 2000

By 

M. Moffatt (Auckland, New Zealand) - See all my reviews
(REAL NAME)   

This review is from: Fatal Defect: Chasing Killer Computer Bugs (Paperback)

Ivars Peterson has written a well thought-out and interesting book that provides just the right amount of depth to this fascinating subject. Both the layman and the software engineer will find Fatal Defect interesting reading.

This book should be required reading for everybody in the IT industry!