August 30, 2010

Measuring up to metric and imperial systems.

No really, metric is interesting. Returning to America has reminded me of the absurdities of imperial measures in everyday life, but also that metric enthusiasts overstate the benefits of metric. Although Thomas Jefferson proposed a decimal system for America in the 1790s, I don't doubt that if further metrication was proposed now it would be condemned as a foreign invention, and contrary to American tradition.

When I first moved to America I was irritated by silly imperial measurements, but more outraged by silly international students who, though intelligent enough to be studying abroad, could not comprehend alternative measurement systems. Being a sometime contrarian I found more virtues in the imperial measures than I first suspected.

Not the least of these virtues is that there are fewer miles in a marathon than there are kilometres. Of course at the same speed the miles take longer to pass. But mentally I find a marathon would be best measured in miles to 20 miles/10km to go, and then in kilometres. After 20mi/32km time is appearing to pass so slowly in most marathons that the achievement of an extra 0.6mi/1km is good to know about.

Other virtues can be found in the temperature scales. When you get below freezing it's nice to still be referring to positive numbers. This saves words, you don't have to say "negative eight," you say twenty. It also allows you to push back the point of being expletively-cold to below zero in fahrenheit, which is about right.

The point is that when you're only thinking about one measure it doesn't matter whether that is metric or imperial. So long as you know the practical implications of the distance or the temperature it doesn't actually matter what scale it is measured on.

Instead, the absurdities of the imperial system are when you have to convert between scales, and when you are relating two measures to each other. Lets take the first of these issues. Ounces and pounds are common measures of weight, and there are 16 ounces in a pound. OK so far, though base 16 is not the most intuitive (base 12 makes more sense than many realize). The problem comes in the store, where some items are weighed in decimal amounts of pounds rather than whole ounces. The conversion isn't obvious between scales when you have some cheese measured at 0.593 pounds and you recipe called for 10 ounces.

It's worse for volume and liquid, there are multiple measures -- fluid ounces, cups, pints, quarts and gallons. Quite why it is sensible to have 10 fluid ounces in a cup, 20 in a pint, and then (then!) switch to 2 pints in a quart and 4 quarts in a gallon. It's not intuitive at all.

The chances of changing this mess are small, since changing United States institutions is difficult at the best of times. Of course the costs of not-ideal measurement systems are small every time you use them, but add up over time and across society. They are probably worth the transition costs, but it will never happen.

Posted by eroberts at August 30, 2010 3:04 PM