A glimmer of hope

[mklotz]

A long time ago I wrote the following broadside. Whenever the metric vs. inferial discussion arises, I like to throw it up so that we all know what the salient points are and which arguments should be avoided.


SOME THINGS YOU SHOULD THINK ABOUT BEFORE ATTEMPTING TO DISCUSS MEASUREMENT SYSTEMS INTELLIGENTLY

The main advantage of the metric system is NOT just that its units are based on powers of ten. Briefly, the key advantages are:

Only one unit used to measure each fundamental or derived physical quantity.

These fundamental units are, as much as possible, simply related to each other, e.g., a liter of water weighs a kilogram.

All subdivisions of fundamental units are related to the fundamental by the SAME sequence of prefixes, e.g., micro, milli, kilo, mega..., each of which denotes a power of ten multiplier.

Derived units of convenience are simply related to the fundamental units, e.g., a hectare = 10000 m^2, a Newton = 1 kg-m/sec^2, 1 liter = 0.001 m^3, etc..

As much as possible, the standards which define the fundamentals are based on phenomena that can be accurately duplicated in laboratories around the world, e.g. the meter is defined in terms of a certain wavelength of light rather than the distance between marks on a bar kept in Paris. Thus the inevitable errors incurred in transferring a standard can be avoided.

Just because something uses a powers of ten sequence does not automatically make it metric. Your inch micrometer has markings for tenths and thousandths of inches but it isn't metric. Most money is subdivided into a hundred "pence" of some sort but it isn't metric.

The fact that something is expressed in metric units does not make it part of the METRIC MEASUREMENT SYSTEM. European thread standards are naturally expressed in metric units because they originate from nations using that system but you will find no mention of thread standards in the documents defining the metric measurement system. If you have a problem with metric threads, take it up with the agencies that set thread standards; don't use it as an excuse to condemn the metric system.

Some folks want to argue that the metric system is flawed because the length standard it uses is "wrong". The French set out to make the meter one ten-millionth of the distance from the Earth's equator to the North Pole. Their technique for measuring this quantity was inspired and they did an admirable job considering the tools available to them at the time. Still, the value they obtained was very slightly in error. But, NONE OF THIS MATTERS IN THE LEAST. All measurement systems are based on a choice of some arbitrary standard. Whether it's the length of the King's arms or the distance traveled by light in a vacuum in a given period of time, it's the fact that everyone agrees to use that length that's important and the real test of a measurement system is how well it satisfies the criteria outlined at the beginning of this text. THE NOTION OF "ACCURACY" OF A MEASUREMENT SYSTEM IS COMPLETE NONSENSE. Accuracy is a function of the measurement tools and techniques, not the system in which the measurements are expressed.

Another argument to avoid is the "we put a man on the moon using..." farce. Everything was built using the measurement system in force at the time. The ancient Egyptians built the pyramids using the cubit (and had problems with multiple, competing definitions) but that's not an argument for returning to their system. If you doubt me, try working a problem using Egyptian fractions. Building achievements are the product of the genius of their designers and fabricators, not the markings on their rulers, scales and buckets. The fact that you managed to build a corral for your dinosaur using the inferial* system is not a valid argument against the use of the metric system.

An insidious so-called argument to avoid is the exceedingly narrow-minded idea that only the aspect of a measurement system you personally use is a proper basis for deciding which measurement system is best for the society at large. It goes something like this...

I don't need another system to machine my parts. The difference between inferial* and metric is just a factor of 2.54, why bother changing; I can make accurate parts using my present inferial system.

There might be some validity to this approach IF THE JOB OF EVERYONE IN THE SOCIETY INVOLVED NOTHING MORE THAN MAKING LINEAR MEASUREMENTS. People have many jobs involving all the aspects of the measurement system. Using one that's antiquated, un-necessarily complex and confusing leads to inefficiency, lost time and dangerous mistakes.

-------------

Nomenclature systems are not a part of measurement systems but they deserve mention here. A sensible nomenclature system should satisfy the following criteria...

* Should provide information about the thing being named. A #43 or size 'K' drill carries no information about its most important feature, the size hole it will drill. This requirement pretty much means that things be labeled by their size in the measurement system in use. A 6 x 1 metric screw tells you OD and pitch directly; a 6-40 inferial* screw makes you work to get the same information. Labeling wire with a number corresponding to how often it's been through the drawing dies may be useful on the wire mill floor but it should never be let loose in the real world where people only care about its diameter.

Not only are sheet metal gauge numbers meaningless, but there are a prolific number of standards to further muddy the waters. Perhaps the USA is finally waking up to the nonsense it's created. In a recent statement, the ASTM has said,

"The use of gauge number is discouraged as being an archaic term of limited usefulness not having general agreement on meaning." (Specification ASTM A480-10a)

* Should have an intuitive progression. Smaller names should correlate with smaller entries in the progression; larger names with larger entries. A #80 drill should be larger, not smaller, than a #1. Similarly with wire and sheet metal.

* Should be open-ended so that a new item larger or smaller than the original set can be sensibly named. This avoids the idiocy of things like 000-120 screws and AAAA batteries. What do you do if you want to add a drill slightly larger than a 'Z'? Again, labeling by actual size avoids most of these problems.

--

* I live in the USA, the only industrial nation on earth that hasn't adopted the metric system. Usonians will tell you that we use the Imperial system but that isn't strictly true. True, we obtained most of our system from Britain but we further botched and confused it into the mess now known as American Customary Units. That's a mouthful to say or write and few will know what you're talking about.

I decided to make up my own designation for a system that has seven different units all called "barrel". My "inferial" is a portmanteau of "infernal/inferior/infuriating" and "Imperial".