Should the United States fully adopt the metric system?

[angus]

I think it already has.  Long ago, in fact.  In 1893 all units, including those English units commonly used in the United States were defined using the fundamental units of the SI.  An inch, for example, is defined very precisely in the United States as 2.54 cm.  Moreover, the SI units are taught in all public and private schools in the United States.  I voted yes, but the adoption was a fait accompli long before any of us were born.

Will we ever have laws requiring bananas to be sold by the kilo?  Will a law ever be passed requiring me to set my dive computer to Pascals instead of psi?  One hopes not.  Not that I have anything against using kilograms, Pascals, nanometers, Joules, etc., and in fact I use those units exclusively at work, and at home when they're more convenient than English units, and when I'm on vacation I'm perfectly comfortable hearing the outside temperature reported in degrees Celcius--which, by the way, also not the SI unit of temperature--but I'm generally against unnecessary bureaucracy.  The free market can and will decide these issues.

No, Sanchez, it's not about forcing anyone to adopt any standards.  It's precisely the opposite of that mentality that my comments are based.

You have to be reasonable.  One is a nice number.  If it makes sense to say something is one centimeter tall, then describe it as one centimeter tall rather than, say, 0.39 inches, or ten thousand microns, or one one-hundredth of a meter.  Just say "one centimeter"  Similarly, if something is about one foot tall, then it's easier to just say that it's "one foot" tall rather than something more cumbersome.  One light year is called "one light year" because it's so much easier than calling it "9.46 × 10^15 meters."  Always use "one" when possible, and let your unit conform to that, otherwise you're just creating unnecessary burden.  Sometimes it isn't possible, of course.  "2.7 Joules" for example is hard to turn into a "one" since it equals 0.0026 BTU, or 0.657 calories, etc.  But don't throw the baby out with the bath water.  

(By the way, the density of water at its densest is one pound per pint, just as it is one kilogram per liter; and the specific heat capacity of water is one BTU per pound per degree Fahrenheit, just as it is one calorie per gram per degree celcius; so it's not like any system has the monopoly on the concept of "one.")

[angus]

It would be nice if we'd gone with a metric inch of 25mm (or a metric mile of 16hm), a metric quart of 1L, and a metric pound of 5dg.

As for temperature, neither Celsius not Farenheit is all that good a choice.  I'd like to either adopt a temperature scale in which the triple point of water was 1000 degrees (and absolute zero was 0). Or one in which ideal gas constant was 1 (which for metric would require degrees about one-eighth the size of what we use now.

How's that?   Nature just doesn't work that way.  Well, you can use Atomic Units (au) as much as possible when writing molecular trajectory code.  Everybody does that, but in the end you have to convert it to some recognizable unit.

Yeah, I agree that having a 1 everywhere is great, and I have put a good deal of thought into coming up with such a system, but as far as I can tell it's impossible.  The SI unit of heat, energy, and work, for example, is the Joule.  It's very simple, but that's because it has a dimensional definition:  1 J = 1 kilogram times 1 square-meter divided by 1 square-second.  The common unit of enthalpy--used not only in the US, but also in China, Europe, and elsewhere--is the calorie (and we can't even get agreement there, because a nutritionist's calorie is one thousand thermodynamic calories.)  Of course, the calorie has a nice thermochemical definition, and you end up with a 1, but it's by design, and highly arbitrary.  Still, 1 calorie = 4.18 Joules, any day of the week.  

As to the gas constant, I actually have this conversation with people regularly:  Convert everything to SI units when possible.  If the problem says, "Suppose you are pumping air into a flat tire.  Suppose that the tire has an initial temperature of 25 degrees Celcius.  Treat the expansion as a reversible adiabatic expansion of an ideal gas.  If the initial volume is three liters and the final volume is 20 liters, what will be the final temperature?"  Okay, that's easy enough (assuming you remember the ratio of heat capacities).  I would advise on that problem to convert everything immediately to SI units:  Initial T = 298 Kelvins, and use the gas constant in units of Joules per mole per Kelvins.  No problem.  

In fact, as a rule of thumb, it's good to convert everything to SI units.  After all, a cubic meter times a Pascal equals a Joule, so there's none of that silly Liter.Atmosphere stuff to mess you up on gas expansion problems.  Punching in a bunch of numbers on a calculator under duress on a timed exam is stressful enough without lots of unnecessary conversions, but the problem is that the real world doesn't work that way.  Neither here nor in China nor in Germany.  There are all sorts of units floating around.  (When was the last time you watched a weather report, anywhere in the world, when the daily maximum temperature was quoted in Kelvins?!)  I did two post-doctoral fellowships, one in California and one in Amsterdam, and I can assure you that the spectroscopists, laser jocks, and engineers in Amsterdam know exactly what a 9/16-inch wrench looks like--mostly because the company that makes the best vacuum fittings is in California, so yeah I recognize that it's a circular argument--but I also am aware that the spectroscopists there use wavenumbers, just like we do, so you have to be prepared to be flexible with the unit of energy for example.  

In any case, if a student gives you the answer in atmospheres, and you have to do a little work to figure out how many kilopascals or bars or psi or mmHg or inches of mercury that is, then it's incumbent upon you to do so.  Mostly just tell them to be sure that they include the units.  1.7 may be the right answer, but if you don't write the units, you will not get the full credit.  I encourage SI units, but you can use whatever units you want.

As for travel, I've walked on the streets of Shanghai when it's 40 degrees--it's often 40 degrees in Shanghai in the summer, believe it or not--and I know that in China 40 degrees is damn hot.  I've also walked on the streets of New York in the winter when it's 40 degrees, and I know that in New York 40 degrees means cold.  In those instances, no units are necessary.  You just have to take it in context.  If you're in the USA, 40 means put on a sweater.  If you're in China, 40 means wear as little as you can legally get away with, and wear a wide-brimmed hat.  In that case, the unit is implied, and needs no explanation.

What we do here works well.  A one-liter bottle of booze is referred to a one-liter bottle of booze, and booze is always sold by the liter or milliliter.  A 12-ounce can of coca cola is called a 12-ounce can of coca cola.  That makes sense as well.  You could, of course, call it a 355 mL can, but think about that:  three hundred fifty-five milliliters has eleven syllables.  12-ounce has two.  I ask you:  which makes more sense?  

Keep it simple, whenever possible.  No need for arbitrary rules that encumber us.


edit:  just for fun, check out this wikipedia article:

http://en.wikipedia.org/wiki/Gas_constant

It gives the gas constant in 27 different units.  I have a IUPAC table in my office that gives it in 59 different units! 

[angus]

So there are always 10 centi____s in a deci____. There are always 1000 milli____s in a deca____, there are always 1,000,000 milli____s in a kilo____. It doesn't matter what base unit you use, the system is always the same.

Are we third graders?  If that's the case, then we already have "fully adopted the metric system" because that's exactly what they're teaching my son.  As far as that goes, that's exactly what they taught me in 1975.  

As I pointed out, we have adopted this system long ago.  Moreover, the US government long ago has agreed to define all units of commerce in terms of high-precision, reproducible phenomena.  NIST.gov keeps a huge catalog of such data.  When it works, anyway.  I posted elsewhere about the hardship it created for me when the website was down because all "non-essential" government functions were closed due to a political squabble.  Where we differ from China and other more authoritarian governments is that we don't have a mechanism whereby the people will suddenly accept a government that tells the people the units in which goods should be bought and sold.  Nor should we, in my opinion.  A five-gallon bucket is more easily described as a five gallon bucket than a 0.0189=cubic meter bucket.  That will be the same in your country as mine.  Don't believe me?  Go to your local hardware store and ask them whether they have 0.0189 cubic meter trash bags.  (To be fair, your gallon is 4.535 liters whereas mine is 3.786 liters.  You also probably spell liters in a strange way.  Still, even though it usually has its head up its ass, my government has a pretty good track record of developing standards, and defines a gallon exactly and our schools teach their students how to properly spell the word "liter."  So long as your country teaches its citizens how its more culturally and economically important neighbors quantify and qualify these units, I can think of no major difficulties these differences in spellings or definitions would create for reasonably well-educated people.  Can you?)

Here in the USA, we really have bigger problems to deal with.  For example, consider the obesity rate.  In any unit, it's a big problem.  To make the problem worse, the nutritional authorities have decided to quantify it with the ridiculous "body-mass index" or BMI.  Can you think of a more stupid quantifier?  It's basically the ratio of a body's mass to the square of its height.  It has units of kg per square meter, but only because it's generally reported as the mass, in kilograms, divided by the height-squared, in square kilograms.  How bizarre is that?  It would be just as stupid if it got reported in units of pounds per square feet.  We obsess over things like "what unit should it be reported in?" in stead of "how useful is this concept?"  It doesn't take into account the differences in density of adipose tissue and muscle mass, or between bone and muscle.  It doesn't account for body types and other differences.  Medical, political, and economic experts have all suggested that it should be refined, if not replaced.

Let's be honest:  unless you are completely devoid of quantitative ability, the units in which quantities are reported are not so important as the relevance of the quantity you are reporting in the first place.  Sure, I'll grant that conversions are easier to do in your head if they're based on some reasonable divisor (10 in the SI, for example), but any unit is fine so long as you report that unit.  If you go out and make a bunch of irrelevant measurements, it really doesn't matter to me what units you report them in, they're still useless.  If, on the other hand, you make precise and useful measurements, then you can report them in whatever units you prefer.  You can even use Klingon units, so long as you define those Klingon units somewhere along the way.

I'll end my involvement in this assinine thread with an anecdote.  On the Harvard bridge, which is actually nowhere near Harvard, but crosses the Charles River from Boston into Cambridge right between MIT and BU, there are markings in units of Smoots.  1 Smoot, 2 Smoots, 3 Smoots, etc., all the way up to 364 Smoots.  The unit is named after Oliver R. Smoot, a fraternity pledge to Lambda Chi Alpha, who in October 1958 lay on the Harvard Bridge and had his frat brothers mark his body lengths.  I graduated from BU, not from MIT, but I hung out at MIT because the other three members of my pop/rock band happened to be MIT students--they all had shorter hair than I did, but they were fun to hang out with and decent musicians, and pretty smart--and I learned quite a bit about the MIT cult from them.  The "smoot" unit started as a frat prank, but it has become an icon of what is arguably the world's pre-eminent center of technical and engineering academic research.  The google unit calculator has even incorporated the smoot.  Physics professors take students out to this bridge to show them the smoots in order to illustrate the concept of non-standard units.  Conversions are made, and epiphanies are in evidence.  There's nothing wrong with the smoot.  There is also nothing wrong with the light year or the foot or the parsec or even the hand, so long as you clearly, carefully, and precisely define the unit relative to a reproducible quantity.

[angus]

     In an ideal world, yes. In reality, it would be difficult for many people to adjust to kilometers instead of miles and kilograms instead of pounds.

Hmmm.  I couldn't resist responding.

First, I can't say that I think this is problematic.  On my second visit to China, for example, I knew that the scale in the pediatrician's office was calibrated in kilograms.  I knew that my son should be about 35 kg.  I knew that if he's only at 32, something is wrong.  When I get back home, I know that the scale in the pediatrician's office weighs in pounds.  I know what his weight in pounds should be and the number of pounds that he might lose from skipping a meal or playing too hard, and the number of pounds that would worry me.  Similarly, I know that when the Japanese authorities board my plane at Narita International Airport with their space suits and high-tech infrared temperature probes to see if the nasty yanks need to be quarantined, they pass out placards in English and Japanese announcing that anyone who has a temperature greater than 38 degrees will be required to stay at a special "hotel" for a few days before being allowed to wander the streets of Tokyo.  It's the over-reaction by the authorities that causes stress, but not the fact that they're using centigrade units of temperature. 

It works the other way as well.  I've had foreign visitors who at first are a little confused, but by the end of the week they know what it means to "drive 75" or "weigh 150" or have "a fever of 102."  I had a visitor from Amsterdam who worked in my lab for a few months and on his first trip to the coffee/snack room he noticed the little spigot that read "Caution:  190° water.  Very hot."  He was astonished!  Then, after a few seconds he smiled.  A little epiphany occurred, and he understood quickly that the spigot did not contain some special super-heated liquid water, but rather water heated to very near its boiling point.  What he found more bizarre, apparently, was the use of fractions on the streetsigns instead of decimals.  He quickly got used to miles but he never really did get used to "Exit 1/2 mile" or "Exit 3/4 mile."  Apparently the use of fractions on the signs really bothered him more than the use of a different unit.

I'm sure that you regularly measure volumes in milliliters when you're working in the lab, and perhaps distances in meters.  We all do.  However, it doesn't cause confusion for you when you have to decide whether you want a photo enlarged to, say, 8 by 10, or 11 by 14 (inches). 

Kilometers is different issue.  It has four syllables.  I avoid using the word whenever possible.  "thousandmeters" has the same number of syllables and is more phonetically pleasing, in my opinion.  If I'm discussing the ozone layer or the tropopause or something like that, I always say "about 20 thousand meters above mean sea level" rather than "about 20 kilometers above mean sea level."  Of course, if I'm renting a car in Mexico, I get into the habit of asking "kilometraje ilimitado?" and the like, but I regard that more a function of thinking in Spanish rather than thinking in the metric system.

I do wish everyone would switch to the SI unit of temperature, because you know you're going to need to convert everything to Kelvins to apply thermodynamic laws, even qualitatively.  As far as I know, no TV or radio weather reporter uses it.  In some places they use Fahrenheit and in others Celcius.  Celcius has the major advantage that it's easier to convert to Kelvins in your head.  Fahrenheit has the minor advantage that its gradations are much closer, so you get a finer scale.  As Ernest pointed out, an even finer scale would be better, but we could just use millikelvins, microkelvins, or nanokelvins if necessary.  Of course Rankine units also allow you to use thermodynamic laws, and will allow you to skip the step of conversion from Fahrenheit to Celcius.  For example, if you're thinking about Charles' law, and you're on a beach far away from a calculator and you just need to do some math in your head--of course, if you're starting with Celcius, then it makes sense to use Kelvins for the application--but if you're starting with Fahrenheit, why not go into Rankine for the application?   

I really think y'all are overthinking this.  It's a simple matter.  Here are salient points to consider:

(1) The US has already adopted the international system, and all of our common units are based on precise definitions. 

(2)  Any system of measurement is a "metric system" because by definition a metric system is a collection of units for measurement, so it's not like there's the metric system.  The English units form a metric system as well. 

(3)  There are practical advantages (and disadvantages) of having multiple units in use, but the advantages of being able to choose which is more convenient outweigh any perceived disadvantages. 

(4)  Certainly children in school should be taught to use SI units of measure, because they will certainly encounter them in the world, but we should not neglect their education in English units as well, for they will encounter them as well. 

(5)  International commerce has been pointed out, but international commerce has not been impeded by different units.  If we can convert dollars to drachmas, then we can convert tons to tonnes.  For example, oil prices are and always has been quoted in U.S. Dollars per barrel, no matter who is doing the buying and who is doing the selling.  The very non-standard unit of the barrel (42 US gallons, or 35 imperial gallons) has become the standard, and there's no need to convert it to cubic meters or liters or anything else. 

(6)  It is the reproducibility (precision) and the accuracy (quality) of a measurement that matters most, not the unit in which those measurements are reported.  So long as you actually write the units on the report, and so long as those units are understood by the readers of your report, then you should be free to use whichever units you find convenient.

[angus]

I really think y'all are overthinking this.  It's a simple matter.  Here are salient points to consider:

5000 words and you missed the most important point.  If you know how much liquid is in a 750ML bottle of Grey Goose, why in the f[inks] would you have trouble with the metric system?  

Ha!  I guess my posts really are too long to read.  In my second post I mentioned that booze was sold by the milliliter.  

Making the standard wine, vodka, and booze bottles 750 mL makes sense because they were previously a fifth of a gallon, which is 757 mL.  They could charge the same price and sell less product, thereby increasing profit.  Liquor distributors had no problem making that change long ago.  The 750-mL bottle has been around for as long as I can remember.

The harder change for them was to stop making quarts (946 mL) and start making liters (1000 mL) because they'd either charge the same price (and gain less revenue) or have to raise the price and piss off customers.  Eventually they opted for the 1.75-liter bottle to replace the half gallon (1.89 liters).  Charged the same price for a 1.75 liter as they did for a half gallon.  Problem solved.

FWIW, just yesterday I got a 750-mL bottle of tequila and a 750-mL bottle of cointreau and a 325-gram container of coarse salt at the liquor store.  Then I went to the supermarket next door and bought five limes (sold at 20 cents per lime rather than by mass) and a small, 4-fluid ounce bottle of lime juice.  Gonna be a margarita weekend.  I have a stack of 35 statistical thermodynamics exams to grade, and I'm sure I'll be thinking plenty about Joules and Pascals and Kelvins and grand canonical partition functions.  Wouldn't want to be without a decent stock of booze to cure that headache.