*totally different*decimal system of metrology, now called SI.

As someone who lives in a country where only SI is used, I can tell you that simply ain't so. And I'm not talking about the survival of named units now sporting rounded decimal metric equivalents, such as the French

*livre*of 500 grams. Those are very few, and fading, and in my country I can recall just one single instance of those (the dunam, an Ottoman unit of area now rounded to 1000 mÂ², or a decare).

Customary units as they used to exist are gone, but the customary way of thinking about weights and measurements is still with us. Even in the most staunchly metric countries. It isn't obvious, which is why it took me a reading about Australia's conversion to SI to realise it. The United States Metric Association website has this to say about the metrication of Australia:

It's not just an insight on a country converting to SI at the time of writing, it's about all of them, including those that converted long ago, like mine. The observation above is one I know to be true from my everyday life.Adult education classes on the metric system failed to attract interest, which confirmed the Board's belief that such courses were unnecessary. It also confirmed the Board's belief that people do not perceive metric in systematic form but learn each unit and its application as an independent and unrelated piece of information. As a consequence the highly logical nature of the metric system or the unsystematic nature of the imperial system had very little meaning for the ordinary citizen.

Of all the SI length units, with or without prefixes, I and most of my fellow countrymen use these: the metre, the kilometre, the centimetre and the millimetre. That's it. Nothing else. No bigger prefixes, and no smaller ones. Gaps too (hecto-, deca-, deci-) are left.

And there's more to be said on that:

SI advocates will tell you you have only to multiply or divide by some power of ten to get an idea of the units. That is, if you just manage to make the metre meaningful to you, all the others are nothing but a decimal multiplication or division away from being so too.

Wrong! I don't divide the metre by a thousand to picture the millimetre, or the opposite to visualise the kilometre. Everyone in the SI-using world takes the units separately, visualising them by the uses they put them to:

- The metre is the height of the shelf in my kitchen.
- The kilometre is the distance of a city-wide walk where I live.
- The centimetre is the length of a single square of chocolate.
- The millimetre is the difference between hair cropped short and a buzz cut.

And it's not just in everyday life. Stating out the distance from Rome to New York isn't an everyday activity, yet the lifelong users of SI stick to the customary unit of long distances: the kilometre. That's despite the fact that such distances span

*thousands*of kilometres, inviting the use of the prefix 'mega-'. But nobody uses megametres (Mm), except for the very few who wish to make a point; it's thousands of kilometres for everyone, including passenger plane pilots. The kilometre is the metric mile.

Or take astronomy. The entire size of the universe could be measured in yottametres ('yotta-' meaning ten to the two-dozenth power), and therefore yottametres, zettametres and exametres could be substituted for all those

*light-years*. Fat chance: light-years are used all the time, as are Astronomic Units, multiples of the distance between our planet and our star. Apparently astronomers think light-years convey the information better than units taken from a system meant 'for all people, for all time'.

But, you protest, scientists working on the microscopic level do use femtometres and the like. That's true. However, this underscores another point in contrast to SI advocacy:

By using SI, say its advocates, one can benefit from having the same units in everyday life as in the science lab. Indeed that would be a great benefit, if it were so. Unfortunately, this theoretical advantage comes up against the fact I mentioned above, that merely having prefixed units does not a bevy of easily visualised units make. Taking a scientist's paper and just reading the measures, I can understand whatever is given in metres or millimetres, but micrometres are already a problem. I have a vague idea of what a micrometre is from discussion of font sizes in printing. Nanometres, well, another hazy idea from the term 'nanotechnology'. Go down to picometres and further and I'm stumped. I have no idea what a femtometre is. You could tell me, but it would still be meaningless for lack of use. It's meaningful to scientists because they use femtometres, just as the demibushel is probably meaningful for American farmers.

That's the point. When SI advocates taunt users of USC about the 'myriad of conversion factors' between units like the firkin, kilderkin, hogshead and demibushel, they fail to realise their system is equally opaque to most people. Yes, there's a nice decimal power relationship between the metre and the femtometre, but it's of no help. Nobody can visualise a quantity by thinking about it as a millionth of another quantity. A tenfold relationship is good; a hundredfold, already vague; a thousandfold gives a very faint idea; and beyond that, those are just numbers, with a train of meaningless zeroes.

The world has mostly given up on its various miles, yards, feet, inches and lines, having non-decimal factors, to replace them with the decimal customary units of kilometres, metres, centimetres and millimetres. The changes are as follows: prefixes instead of unrelated names, consistently decimal relationships, and worldwide coverage. The first change might be easier on the memory. The second change is actually a disadvantage (because, as we all know, dividing by three is a dicey proposition in base ten) and in many places not even a change (Chinese weights and measures have always had decimal relationships).

The third change is the only substantial advantage of the worldwide conversion to SI. Truly it was easier for late 19th-century Italy and Germany to pick up the deci-metric system instead of deciding on one of so many local pre-unification standards, and it was a blessing to be rid of the plethora of weights and measures extant in the Ottoman Empire. And it is easier today when weights and measures mean the same to all people everywhere. But this is a benefit of

*standardisation*, not an intrinsic one; the virtue being pushed here is that of conformism, and it would be valid no matter the system used. The debate on the intrinsic merits of SI and other system is something else entirely.

In conclusion, the old customary units themselves could be abolished, but the customary

*way*of using weights and measures,

*any*weights and measures, endures because it is the way humans do it. Seen in this light, the conversion to SI is not such a great change as it is portrayed; and it is all the more lamentable that old customary units often having useful factors between them have been replaced by new customary units related by factors of ten alone. The scientist and the non-scientist have not been brought closer, and the needs of practical life have been set back by the adoption of a system that confines all numeric representation of weights and measures to base ten, providing next to no escape routes for more useful divisions.

TGM beckons.