have made it the most powerful naval vessel of the time, capable of delivering a broadside of devastating proportions. The men he had contracted to build his ships attempted to explain that the ship had too little ballast to support two gun decks, and that the resulting ship likely would be unsafe to sail. The King insisted—just like, say, many project managers—that his orders should be followed. On a software project you can quit, but if the King is your boss you might lose more than your job—you might, say, lose your head—so the project went forward.

In 1628 the ship was finally ready for quality assurance (QA) testing. Seven-teenth-century QA of ships was a bit different from what might happen today. Thirty sailors were picked and asked to run back and forth, port to starboard, across the deck of the ship. If the ship didn’t tip over and sink, then the ship passed the test. You did not want to be on the QA team in 1628. After only three runs across the deck the Vasa began to tilt wildly and the test was canceled. The test may have been canceled, but not the project. This was the King’s ship, after

all, and she would sail. And sail she did.

On August 10, 1628, in a light breeze, the Vasa set sail. She was less than a mile from dock when a stiff breeze knocked her sideways. She took on water, and sank in full view of a crowd of thousands of onlookers. Approximately 30 to 50 sailors were killed when they were either trapped in the ship or were unable to swim to shore.

In response to the catastrophe, the King wrote a letter insisting that incompetence had been the reason for the disaster. He was, of course, correct, but not in the way he might have envisioned. An inquest was held and the surviving members of the crew, the captain, and the ship builders were questioned as to the state of the crew and the ship at the time of the incident. The mostly unstated belief by the end of the inquest was that the design had been a failure and the designer had not listened to the builders about the shortcomings of the design. Of course, the King could not be held at fault, so the final verdict was an “act of God.” As a related aside, the disaster was also a huge economic loss for Sweden.

Now, this story may not be as well written as Frankenstein, but it’s a much more direct warning about engineering failures. I think the funniest or saddest part of this story is how modern it is. Nothing has changed since 1628. People still fail to communicate, leading to failures of disastrous proportions. Egos get in the way, mysterious supernatural forces are blamed for human failings. It’s all kind of obvious in a really sad way.

In the 1960s the Vasa was raised from the bottom of the bay in which it had sunk and eventually placed in a museum in Stockholm. I visited the Vasa in 2000 as part of the SIGCOMM conference. The whole story is told there in the plaques on the walls. It’s a museum all engineers ought to visit at least once.

KV

George V. Neville-Neil ( [e-mail]) is the proprietor of Neville-Neil Consulting and a member of the ACM Queue Editorial Board. He works on networking and operating systems code for fun and profit, teaches courses on various programming-related subjects, and encourages your comments, quips, and code snips pertaining to his Communications column.

ACM Journal on Emerging Technologies

i n Computing Systems

This quarterly publication provides comprehensive coverage of innovative work in the speci cation, design analysis, simulation, veri cation, testing, and evaluation of computing systems constructed out of emerging technologies and advanced semiconductors. Topics include, but are not limited to: Logic Primitive Design and Synthesis; System-Level Speci cation, Design and Synthesis; Software-Level Speci cation, Design and Synthesis; and Mixed-Technology Systems.

http://jetc.acm.org/