Technology and accidents

Several recent highly publicized plane crashes, along with the accident on the D.C. subway system (Metro), have got me thinking, albeit not very coherently, about accidents and technology. The Metro accident hits closest to home for me. Was it the kind of thing bound to occur sooner or later given the (presumed) pitfalls of reliance on an automated computerized system, or could it have been prevented by, say, a back-up system? Such a back-up would no doubt have been expensive, but I believe airplanes have them, at least in a rudimentary form, and so do, I believe, nuclear power plants, so why shouldn't subways? Perrow's work on 'normal accidents', which I've heard about but not read, suggests that not all accidents can be eliminated, especially in technologically complex environments; still, that is no reason not to try to make transport and other systems as safe as possible.

The Metro accident is a tragedy, obviously, for those who were killed and injured and for their families; it also will inconvenience everyone who uses, either regularly or occasionally, the Red line, as service on that line will be slower than usual. Moreover, because the whole Metro system is going to be operating in manual (non-automatic) mode for an indefinite period (at least according to what I heard last night on the news), the service on the system as a whole will be slower. Of course, safety matters more than speed, but in a metropolitan area already choking on its traffic -- an area where one can easily get the impression that no one does anything except drive around all day and clog up the roads -- anything that makes the subway less fast and efficient is bound to be unwelcome, to say the least.

On a related issue: See this post on the projected high-speed rail line between San Francisco and L.A.