Month: September 2021

As an American and a person with an OCD streak, my country’s use of the Imperial system of weights and measures has long bothered me. The Metric system is simply superior, and it pains and embarrasses me to think about how much my country’s stubbornness hurts global commerce and confuses millions of people the world over each day.

How many car accidents happen because drivers visiting another country confuse kilometers per hour with miles per hour, or vice versa? How much U.S. – foreign trade is made impossible by manufacturers on either side of the border making their products to simple, whole-number measurements (like 12 ounce cans of soda) under their national systems?

The human factor is behind all of this. Like anyone else, Americans find it hard to shake old habits, and the use of the Imperial system is one of them. Humans have an instinctive aversion to change, even if they know in the abstract that a change will benefit them. Moreover, our limited brainpower imposes real hardship to learning new standards of weights and measures. Most American adults who are only used to the Imperial system would, even if forced to use Metric, never fully adapt, and would continue thinking in pounds, miles, and degrees Fahrenheit until the day they died, and would use imperfect heuristics in the interim to guess what the right Metric value was. This phenomenon–in which an old standard becomes fixed in place because the up-front costs of changing it are perceived to be too high even though something superior is known to exist–is called “path dependence.”

But what happens when human stubbornness and human cognitive limitations stop being factors? What happens when technology provides workarounds, or renders different national standards moot, or lowers the costs of changing them to affordable levels? For example, once autonomous cars are ubiquitous and humans don’t drive anymore, what would be the harm in changing all the road signs in America to indicate distances in kilometers and speed limits in kilometers per hour (kph)? Since humans aren’t driving anymore, there’s no longer any potential for Americans to get confused by the units of measurement written on road signs.

And even if the road signs weren’t changed and continued making reference to miles and miles per hour, what would it matter? Human passengers wouldn’t need to look at the signs anyhow, and probably wouldn’t even want to glance at them since it would distract them from watching videos or playing video games on their personal devices.

The consequences of using the Imperial system for weight and volume measurements will also diminish with time. Aside from weighing their own bodies on scales, Americans only think about weight and volume when buying gasoline and groceries. As electric cars become the standard, and as machines do more shopping and at-home food preparation for people, Americans will lose any instinctive grasp of how big a gallon, pound or ounce is, and the machine-run economy will be able to quietly transition to the Metric system without confusing anyone since everyone will be too busy playing VR games or doing recreational drugs to ever think about it.

The American use of the Fahrenheit scale is already not very consequential since people mostly only think about temperature when looking at weather forecasts. People commonly view those forecasts on their smartphones, making it possible for individuals to remain ensconced in their own Fahrenheit or Celsius information bubbles wherever they go, without anyone around them being the wiser. People also have to think about temperature when baking meals, but as I pointed out, machines will take over those tasks in the future, freeing humans from having to think about it. The U.S. could officially switch to the Celsius scale now if it wanted to, with minimal disruption or benefit.

Relatedly, the QWERTY keyboard is an example of a suboptimal technology whose use is locked-in. Alternative keyboards, notably the DVORAK, are better since they place the most commonly-used keys directly under where the typist’s fingers rest, reducing the finger and wrist movements needed to type words. As a result, a person trained to use a DVORAK keyboard can type faster than one trained to use a QWERTY keyboard. Letter frequency distributions across languages that use the Latin alphabet like English, French, and even Turkish, are surprisingly similar to each other, meaning DVORAK or some close derivative of it would be the optimal keyboard for them all.

So why doesn’t the Western world switch to DVORAK keyboards? Blame the human factor again. As the small number of people who have switched from QWERTY to DVORAK can attest, it takes a lot of time and effort to re-train the muscle memory in your fingers to adapt to a new layout. Humans are bad at learning new things and forcing themselves to forget old habits. Not so for robots, who will be able to adapt to any new keyboard layout instantly. In fact, they probably wouldn’t use keyboards at all and would instead directly interface their minds with other machines to wirelessly input commands, as would humans who had computer implants in their brains. In short, even a locked-in device like the QWERTY keyboard will eventually give way to something better because users will get capable enough to switch to new keyboards, or will evolve beyond the need for any type of keyboards. (And in the nearer term, greater use of continuous speech recognition (CSR) and improvements to autocomplete algorithms will also minimize the importance of typing skills.)

Returning to the example of autonomous cars, it’s clear that they would mollify the effects of another inconsistent standard–the side of the road cars have to drive on. Differences between countries stymie trade, as it’s very hard for truck drivers accustomed to driving on one side of the road to mentally switch to doing the opposite once they cross a border, and because cars themselves must have their layouts configured for right- or left-handed driving, adding to costs.

However, autonomous vehicles could, at the flick of a switch, adapt from one driving orientation to another with no problem at all. And since they wouldn’t be designed to accept human commands, the cars wouldn’t have steering wheels, pedals, or a special position for a human driver (there would only be two passenger seats in the front row), meaning an autonomous car meant for sale in, say, Britain, would have the same internal layout as a car meant for sale in France.

Thanks to technology, transnational differences in what sides of the road cars must drive on could persist indefinitely, without imposing any costs or inconvenience. Of course, other inconsistent standards would defy technological workarounds. One that springs to mind is differences in railroad track gauges, which is how far apart the two rails are. Rail cars and locomotives must have their wheels spaced properly to engage with the tracks, so rolling stock configured for one gauge can’t operate on railroad networks using a different gauge. This hurts trade between countries like Russia and China since export cargo has to be unloaded at their border from trains using one country’s gauge to trains using the other gauge.

The only solution to this problem is to physically widen or narrow existing railroads until they are a different gauge, and to make corresponding modifications to the wheels of train cars and locomotives. It can be done, but it’s expensive work, partly because of the disruptions in rail service that happen while everything is being changed. For that reason, very few countries now find it worthwhile to change their railroad gauges.

However, free machine labor will change that calculus in the distant future. Robots will inevitably do every type of work cheaper and better than humans, and will eventually number in the hundreds of millions or even billions. Such an enormous labor force, willing to work for free and without complaint, would make hitherto uneconomical projects possible (I discussed this in my I, Robot review and how it would lead to the construction of a lot of new infrastructure). For example, fifty years from now, if Spain voted to switch to the “Standard Gauge” used by the rest of Europe, the project might last a month and simply entail every household agreeing to loan their robot butlers to the government for a few hours each day to work in teams narrowing nearby segments of railroad or narrowing the wheel distances of train cars. If this sounds silly, realize that something like this was done in the American south in 1886: over just two days, thousands of men across the region changed the gauge of 11,500 miles of track to match what we being used in the northern states.

The lack of a global standard for electrical plugs and outlets is also a drag on commerce, particularly in Europe where there are six different national standards. Not only is it annoying for travelers who find their electronic device plugs don’t fit into wall sockets, it is costly for manufacturers of said devices since they need to made different versions for different countries, or include plug adapters with their products.

Fortunately, there is a Europe-wide standard for mains electricity, meaning the power coming out of all electrical outlets on the continent is 230 volts at 50 hertz. As such, it would be possible for all of Europe to switch to a single type of electric outlet, though this is again foiled by the high costs of replacing several billion outlets (a simple task, I assure you from experience, but very intimidating for someone who has never done it). Again, ubiquitous robot labor could overcome the problem. Every household’s robot butler would count the number of electrical outlets in his house, tell the government, wait to receive a shipment of new electrical outlets, and then install the new outlets over the course of a day or two.

I don’t have the time to discuss every single productivity- and trade-sapping inefficiency that springs from nonuniform or suboptimal standards, so I’ll leave it to you to imagine how technology could overcome them as well in the future. And also know that intelligent machines and our much smarter posthuman descendants will have it within their power not just to sweep aside the inefficiencies we’ve created, but to upgrade to new, global (and truly universal) standards that are more efficient that anything we have conceived of. Gifted with greater powers of rationality and self-control, the intelligent beings of the future could do things like create new languages and alphabets that convey information more efficiently than any we humans have made so far. Thanks to some properties of electricity, there might be a specific voltage / frequency combination that is optimal for electrical grids, but which we haven’t implemented because it’s something weird like 183.04 V / 49.92 hz.

There might even be something better than the Metric system. One of the glories of that system is that it is entirely base-10, whereas the Imperial system’s units don’t follow any consistent scaling pattern (e.g. – there are 8 ounces in 1 cup, 16 ounces in 1 pint, 32 ounces in 1 quart, and 128 ounces in 1 gallon). But base-10 systems of measurement are only useful because humans are terrible at doing math in their heads, and because we have ten fingers to count on. A being with superhuman intelligence could just as easily use the Imperial system, or a system that hasn’t been invented yet that was not base-10 (I happen to think base-12 systems are superior) and maybe had some of the inconsistent aspects of the Imperial system.

Just take heart, fellow OCD people of the world. Our day is coming!

Links:

1. In 1886, American workers using only simple tools changed the width of 11,500 miles of railroad track in two days.
   http://southern.railfan.net/ties/1966/66-8/gauge.html
2. The DVORAK keyboard is better than the QWERTY keyboard.
   https://thekindle3books.com/qwerty-vs-dvorak-the-two-great-keyboards-the-time-were-born/