"Mistakes and 'close enough' are the ways to build bridges that fail."

Quite a while back I excerpted a quote from Purple Avenger that I liked. My post was entitled The Engineer's Perspective. Here's the quote:

My best friend is a lawyer, bright, gifted, ... PhD in law; bored with his job, he decided to study engineering. After his first quarter, he came to me and said that the two "C"s he'd achieved in Engineering Calculus 101 and Engineering Physics 101 were the first two non-A grades he'd ever gotten in college, and that he had had to study harder for them than for any other dozen classes he'd had. "I now understand", he said, "why engineers and their like are so hard to examine, whether on the stand or in a deposition. When they say a thing is possible, they KNOW it is possible, and when they say a thing is not possible, they KNOW it is not. Most people don't understand know in that way; what they know is what we can persuade them to believe. You engineers live in the same world as the rest of us, but you understand that world in a way we never will."

I don't think that you have to love math to be an engineer, but you are going to have to learn it. That means that you're going to have to do the homework, correctly. Mistakes and "close enough" are the ways to build bridges that fail.
htom | 09.26.05 - 2:10 am

This week, a bridge failed. It was not, particularly, an engineering failure. The bridge had stood for some 30 years. It was a management failure.

Let me explain.

Back in March, 2005 I linked for the first time to Dr. Sanity, the blog of Dr. Pat Santy who was a flight surgeon for NASA for the Challenger mission. In that piece I reflected on the effect that the Challenger disaster had on me - at the time, a recent college graduate looking for a job:

I remember listening to the launch of the Challenger early in the morning here in Tucson, and thinking - as the station broke for a commercial - "At least this one didn't blow up on the pad."

Morbid, I know, but I'm also an engineer. I wasn't then - I was still going to college at the time (Ed. note: I actually graduated in December of 2005) - but that's been my orientation for most of my life. I knew that each manned launch was a roll of the dice, a spin of the cylinder in a big game of Russian Roulette, and that NASA had become just another government bureaucracy. (And I also knew just how close we had come to losing three men in Apollo 13 because a series of small, innocuous errors had cascaded into a catastrophic failure in a system that was almost neurotic in its quest for safety.)

It was just a matter of time.

Still, I was shocked when they came back from commercial to announce that Challenger had been destroyed in a launch accident just minutes after liftoff. I knew that all seven of the astronauts were dead. I knew that the "teacher in space" wasn't going to get there, and that a classroom of students had to be devastated by that realization. Many, many classrooms, but one in particular.

I watched the footage of the liftoff, now splayed in endless grisly loops on every network - all of which had previously declined to show the launch live and interrupt really important stuff like "Good Morning America." I watched as the flame bloomed out from a Solid Rocket Booster joint, impinging on the huge external fuel tank, and said, "That's what killed them. What the hell caused that failure?" I watched the Satan's horns of the SRB exhaust tracks as they trailed up and away from the epicenter of the blast. And then I watched it all again.

Over and over.

Later I discovered that the engineers at Morton Thiokol had tried to get the launch scrubbed, knowing the problems that cold weather caused in the O-ring joint seals of the SRBs, but they had been told to "take off their engineer hats and put on their manager hats" in order to make a launch decision. The launch had been delayed too many times, and President Reagan would be making his State of the Union address that night, with a call to Crista McAuliffe - Teacher in Space.

I decided right then that I didn't ever want to be a goddamned manager.

Judging from what's being reported, engineers knew for some time that this bridge, like one in eight around the country, had "structural deficiencies" due to fatigue, corrosion, sub-standard assembly practices, and so forth. This means that there are a lot of bridges (and, one assumes, other infrastructure) out there that aren't up to their design capacities any more.

How bad was the 35W bridge? Apparently pretty bad, but not so bad that some engineer somewhere was willing to risk his job over it. I'm sure that more than one structural engineer was told to "put on his politician's hat" and make a decision based on economics and politics rather than safety.

Bridges fail. But it's more often than not due to non-engineering causes.

UPDATE: (Hat tip to Shooting the Messenger) 'Go after the designer,' says Minneapolis bridge checker. Apparently this weasel wants to avoid the fact that this bridge lasted forty years before it collapsed, and that it was his job to determine if it was still safe, not the designer. The designer did his job. Mr. Kurt Furhman, bridge inspector, probably kept being asked to put on his "politician's hat" - and did so.