American Scientist

To the Editors:

I found Howard Wainer’s Macroscope column on the power of convention “Why Is a Raven Like a Writing Desk?” (November–December 2008) quite enjoyable. I would like to add one of the most depressing examples of convention’s stifling effect: The antiquated system of British Imperial units used in the United States. The metric system clearly is superior: All units are based on precise standards, most of which can be independently recreated in the laboratory. But even more important, in the metric system measures connected by geometrical laws convert easily into each other: a cube of 1 meter side length holds 1 cubic meter of water, a kilometer is equal to 1,000 meters. Quick: how many fluid ounces fit in one cubic foot? How many inches in a mile? These disparate units confuse endless generations of school children. The power of convention and tradition and inertia inherent in the world’s largest economy has stymied attempts to join the rest of the world.

Sebastian Kuhn
Norfolk, Virginia

To the Editors:

Wainer’s points are well taken with respect to graphic representation of data and the QWERTY versus Dvorak keyboards. But English spelling? If it were phonetic, which dialect or accent would rule? Aside from geographical differences (regional, national), there will be temporal differences (historical evolution), for example: “the great vowel shift.” In the early days of written English, a particular regional dialect was preferred; it could just as well have been another.

Could we order up a worldwide reform of English, including a provision that our children pronounce words exactly the same as we (the reformers) decide?

Morton Nadler
Blacksburg, VA

To the Editors:

Wainer in his entertaining article doesn’t explain the perverse endurance of established conventions. The strength of any conventional design is its very familiarity.

For pie charts, unlike dot charts, it is self-evident there is a “whole” divided into its constituent parts (which could also be arranged in order of decreasing size). Where part-whole relationships are significant, the pie chart should have an advantage. About keyboard layout: first, most people don’t type for a living and are satisfied with adequate rather than optimum efficiency; second, even hunt-and-peck typists can leverage their motor memory to advantage when the layout is familiar.

Bill Benson
Kensington, CA

To the Editors:

The pie chart on page 447 (November–December 2008) has organic waste at 37.3 percent and paper waste at 30.8 percent, but the organic section shown is smaller. Oops?

R. F. Trimble
Carbondale, IL

Dr. Wainer responds:

It would be hard to summon an argument for the English system of measurement over the metric. I will not try to do so. But the metric system is not perfect. In measuring human weight, the kilogram is a bit too large and the gram way too small. In this era of obesity, if we waited to diet until after gaining an extra kilo rather than an extra pound, we would lose valuable time. In the same way, one degree Celsius is a tad large, whereas degrees Fahrenheit seem right. Connecting across measures within the English system is not always odious. Consider the ancient “a pint’s a pound the world around” that connects liquid volume and weight accurately and with more charm than 1 cubic centimeter = 1 gram.

The pie’s chief advantage over the dot chart is indeed the visual reinforcement of how all pieces sum to 100 percent. But that’s it. I opted to present dot charts rather than bars for many reasons—a leading one grows from Tufte’s data-ink ratio. The only part of the bar that conveys information is the line marking its end. The rest serves only to hold up that line. A dot chart emphasizes the aspect of the icon conveying information; its data-ink ratio is much higher.

The pie chart that appeared with my essay had an error that unintentionally illustrated another flaw in the form. The version I prepared attempted to duplicate the original from the New York Times that had a pseudo 3-D effect. Thus the segments were proportioned differently to provide perspective. American Scientist editors correctly dismissed the 3-D effect as specious and omitted it. They and I missed that the perspective remained and yielded the error many readers noticed. My apologies.

Last, I purposely omitted a discussion about which pronunciation a standard spelling of English should represent. Such a choice would have to be made.