02 August 2007

The Good Thing About Living In Earthquake Country . . .

. . . is that, for the most part, bridges, overpasses and such are "over" engineered for normal use. It's unknown what the specific causes were for the terrible bridge collapse in Minneapolis (a good starting point for coverage is Buzz.mn, a good place to marvel at the disgusting nature of some people on the internet is the comment section at HuffPo).

Not that our infrastructure is perfect in CA, freeways have buckled, and bridges have collapsed here, too (though I don't recall that happening without a proximal cause such as quake, or tanker fire), failure in all human systems is inevitable, and usually the decision making process that lead to the failure was reasonable, yet somehow flawed. It's hard to fix, and even inspect systems that are under continuous heavy use, yet that's what we do with all sorts of infrastructure. It's not a uniquely American problem, it's a human problem. Stuff happens, often for cause, causes that were foreseeable and obvious in retrospect, but hindsight, as they say, is done by asses.

Seems like the 2006 inspection of this particular bridge will get heavily scrutinized now, but it's likely that despite the problems spotted, there was nothing remarkably different about this bridge than all the other similar spans (which is why they figured it was solid without major review until 2020). Had some engineer suggested shutting down every such span and working every flaw till each was perfect, that engineer would be called a Cassandra, and unrealistic. Yet, most likely some Cassandra will step forward and claim they were prevented from doing just that.

Speaking of "The Logic of Failure", I recommend the book by that title. It's a bit dense, a bit jargon heavy, and it's translated from the German so some of the language gets a bit tortured. But as a short (yet thorough) examination of decision making and the use of computers to model behavior, it's a damn fine read.

Humans are flawed, especially when it comes to assessing long term risks. Certain fallacies crop up again and again in a variety of situations, but the psychology behind these faults are often the same, the book mentioned lays this out in detail. Pick it up at your local library, or order it from Amazon, it shouldn't disappoint.

Also, Tammy Bruce has a good post about not posting about this immediately. There's really nothing to say other than to be thankful if you weren't affected, be thoughtful for the ones that were, and be mindful that now isn't the time for judgement and recrimination.

James Lileks is probably getting traffic to Buzz.mn like never before (most every blogger I've read link him when mentioning this story, for the blogosphere Lileks=Minneapolis and Minneapolis=Lileks, Buzz.mn probably wouldn't have accrued all that benefit had Lileks not taken over 8 weeks ago), and I'm sure he'd trade every last eyeball, and all past and future eyeballs for a way to go back and prevent this from having happened.

(and it's probably true what Douglas Adams said about time travel, the main problem with it is grammatical, though some are attempting to create a framework for discussion)

No comments: