Friday Topsight, September 8, 2006
Just a quick one today, with not as much text -- but good links to hang onto.
Smeed's Law: What happens when you add cars to traffic? The number of accidents goes down. On average, annual increases of traffic volume leads to a decrease in accidents per vehicle. That was the observation of one RJ Smeed in 1949, and the rule has proven true (much to everyone's surprise) time and again.
The Wikipedia entry on Smeed's Law is terse, but an interesting article by Gwynne Dyer sent to me in email gives a bit more discussion:
Around the world, about 1.2 million people are killed in road accidents each year. An astounding 85 percent of those deaths happen in developing countries, although they own less than a fifth of the world's vehicle fleet. [...]
The amount of road traffic in the United States has grown fourteen-fold since 1925. If the number of American deaths per million miles (kilometres) driven had stayed steady at the 1925 rate, there would now be 300,000 deaths per years on American roads, not 40,000. [...]
Smeed offered no explanation for this phenomenon, but I think that there is a collective learning process as more and more people become experienced drivers, and particularly as the generations turn over and children grow up in families that already own cars.
As with all such observations, there is some dispute over its details, but as a general rule, it appears to hold broadly true. Dyer's explanation, the "collective learning process," makes sense, but it suggests to me that there might be an even bigger rule at work.
Is there a Smeed's Law for other technologies? Does the rate of accident or unintended misuse for other technologies go down as the technologies become more commonplace? Put that way, it seems likely, and would come from a combination of better skills, collective learning (e.g., observation while young of appropriate technology use so that it becomes an instinctive behavior), and systems in the technology to help users avoid problems, whether we mean anti-lock brakes or VCRs that automatically set the time (no more blinking 12:00).
Social Footprints: Joel Makower writes about the concept of the Social Footprint of climate change, described by the Center for Sustainable Innovation as the "quantitative measures of the social sustainability of behaviors -- collective organizational behaviors, in particular."
The goal of measuring social footprints, says CSI, is "to assess the sustainability of organizational operations in terms of their impacts on strategies for achieving climate change mitigation."
Translation: the social footprint looks beyond whether a company's efforts to reduce its greenhouse gas emissions are merely worthy or exemplary, but whether they are sufficient to actually solve the problem. [...]
It goes beyond merely tracking aggregate greenhouse emissions, assessing whether a company's climate strategy and performance can actually help stem a climate crisis over the long term.
I like this. I must admit to having some deep hesitation around the "ecological footprint" concept, except as a very broad metaphor; most of the footprint measurement tools involve way too much hand-waving and assumption (around water use, around consumption of non-energy resources, around energy production) for my taste. This metric, conversely, is both more narrow and more justifiably quantitative than the "how many planets are we consuming?" footprint model. In addition, rather than looking at a snapshot of how your behaviors map today, it looks at ongoing dynamics. Green thumbs up here.
Set Kirkyans on Stun: In Wednesday's post on (Virtual) Weapon Smuggling, I mention in a somewhat off-hand way about Sven Johnson's concept of Kirkyans. He returns the favor with a new write-up, linking to the Weapon post, going into more detail about the Kirkyan idea, and an unexpected challenge it may present. In short, the co-evolution of physical manufactured objects and their virtual counterparts could be particularly suited to weapons development.
It seems to me that the Kirkyan model does not need to apply solely to objects per se. A building design that "learns" through virtual iteration could be a Kirkyan. Moreover, if we want to think about the thuggish implications of the physical world/virtual world overlap, there's nothing to stop a group of criminals from testing and perfecting a crime using a virtual model of a real space; in fact, I'd be a bit surprised if it hasn't happened already.
Hmm. I think I may have a movie pitch here.