3 ms·

The graph shows a continuing decline, with a sizable difference between even now and 1990. In fact, almost the entire trend since 1918 has been roughly linear —

by scrod 16y ago

The graph shows a continuing decline, with a sizable difference between even now and 1990. In fact, almost the entire trend since 1918 has been roughly linear — hardly what your hypothesis would suggest.

mattdeboard 16y ago
I don't know about that. Looks to me, just from eyeballing it of course, that there are two periods of logarithmic decay. The first in 1918-1969, then another period of logarithmic decay until present day. A brief Google search reveals some information that would seem to support there was a big change in fire safety in the 1970s: http://berkeley.edu/news/media/releases/2007/08/09_williamson.shtml http://berkeley.edu/news/media/releases/2007/08/09_williamso... >"[Williamson's contributions] helped make the late 1970s the high-water mark of concerted fire safety science research in the United States, a level which has not subsequently been exceeded." These contributions began in the early 1970s beginning with the classifications of plastics and their flammabilities. The overall tone of the article (admittedly, it is an obit for a noted physicist in the fire science field so it's bound to be flattering) indicates that the decade starting in 1970 began a sea change in fire safety and great gains in fire science. That's my point. Big technological advances increase survivability and decrease risk. I don't get the round-eyed, slack-jawed wonder at the fact that there has been a huge drop in fire-related deaths in 89 years. Moving from open flames to electricity then improving building codes etc., provided two huge catalysts to change. Given that drastic technological change was the constant of the 20th century, this Freakonomics post just seems silly. (As another responder said -- there's also been a dramatic decrease in cases of scurvy over the same timeframe. Similarly a "duh" revelation. I bet a graph of cases of kuru and measles would look about like the fire one.)

by scrod 16y ago

mattdeboard 16y ago

I don't know about that. Looks to me, just from eyeballing it of course, that there are two periods of logarithmic decay. The first in 1918-1969, then another period of logarithmic decay until present day. A brief Google search reveals some information that would seem to support there was a big change in fire safety in the 1970s: http://berkeley.edu/news/media/releases/2007/08/09_williamson.shtml http://berkeley.edu/news/media/releases/2007/08/09_williamso... >"[Williamson's contributions] helped make the late 1970s the high-water mark of concerted fire safety science research in the United States, a level which has not subsequently been exceeded." These contributions began in the early 1970s beginning with the classifications of plastics and their flammabilities. The overall tone of the article (admittedly, it is an obit for a noted physicist in the fire science field so it's bound to be flattering) indicates that the decade starting in 1970 began a sea change in fire safety and great gains in fire science. That's my point. Big technological advances increase survivability and decrease risk. I don't get the round-eyed, slack-jawed wonder at the fact that there has been a huge drop in fire-related deaths in 89 years. Moving from open flames to electricity then improving building codes etc., provided two huge catalysts to change. Given that drastic technological change was the constant of the 20th century, this Freakonomics post just seems silly. (As another responder said -- there's also been a dramatic decrease in cases of scurvy over the same timeframe. Similarly a "duh" revelation. I bet a graph of cases of kuru and measles would look about like the fire one.)