Re: SI &c., eh?

[Leigh Palmer <palmer@SFU.CA>]

> 1. Any unit may take only ONE prefix. For example 'millimillimetre'
> is incorrect and should be written as 'micrometre'.

Some of you older types will remember the micromicrofarad, usually
abbreviated mmf (or MMF!) in the devices themselves. The picofarad
(known fondly as the "puff") is a distinct improvement. This rule
also excludes the ever popular ion propulsion thrust unit, the
millimicromousefart.

> 2. Most prefixes which make a unit bigger are written in capital
> letters (M G T etc.), but when they make a unit smaller then lower
> case (m n p etc.) is used. Exceptions to this are the kilo [k] to
> avoid any possible confusion with kelvin [K]; hecto [h]; and deca
> [da] or [dk]

Ah, an excellent rule - only three exceptions? Well, there is the
Greek letter mu, and astrophysical applications could be found for
the teratesla (TT)

> 3. A unit which is named after a person is written all in lower
> case (newton, volt, pascal etc.) when named in full, but starting
> with a capital letter (N V Pa etc.) when abbreviated. An exception
> to this rule is the litre which, if written as a lower case 'l'
> could be mistaken for a '1' (one) and so a capital 'L' is allowed
> as an alternative. It is intended that a single letter will be
> decided upon some time in the future when it becomes clear which
> letter is being favoured most in use.

The period of time necessary for this to happen would be called,
I suppose, the darwin.

Was Georg Simon Ohm Greek? The committee should have considered
changing the symbol to O for uniformity's sake.

I will also note that the fermi is an exception to this rule, and
I believe it predates the femtometer. It symbol is in lower case.

> 4. Units written in abbreviated form are NEVER pluralised. So 'm'
> could always be either 'metre' or 'metres'. 'ms' could represent
> 'metre second' (whatever that is) or, more correctly, 'millisecond'.

Suppose the unit is written out or spoken as a word? Does one
use the plural then? In particular, how should one verbalize the
quantity 0.7 m ? Or 0.0 s ?

> 5. An abbreviation (such as J N g Pa etc.) is NEVER followed by
> a full-stop unless it is the end of a sentence.

It has been my idiosyncratic practice to intercalate a blank
between a sentence-ending-symbol and the period. (I also do odd
things with quotation marks which, I believe, make sentences
easier to parse.) English is tolerant of innovation.

> 6. To make numbers easier to read they may be divided into groups
> of 3 separated by spaces (or half-spaces) but NOT commas.

Exceptions should be tolerated. For example, 1 000 is not a number
that one reads fluently. I maintain that 1,000 is easier to read.
Its only drawback is that in languages other than English it has a
different meaning, a bigger problem than ease in reading.

> 7. The SI preferred way of showing a decimal fraction is to use a
> comma (123,456) to separate the whole number from its fractional
> part. The practice of using a point, as is common in English-
> speaking countries, is acceptable providing only that the point
> is placed ON the line of the bottom edge of the numbers (123.456).

Now that one will never sell! I know of no journal in the English
language which would tolerate commas in that context.

> 8. It will be noted that many units are eponymous, that is they
> are named after persons. This is always someone who was prominent
> in the early work done within the field in which the unit is used.

E.g. John Deere. I'll use one of his units in my field if it ever
dries out. Many good men (and one woman) have had their immortality
terminated by this committee, a sort of post mortem execution. Can
anyone name the famous physicist (NMR and general relativity) whose
unit is now threatened with extinction by metrication?

I hope that you will not be too hard on me for having some fun with
this topic. I think all of this organization shows just how hard it
is for an international committee to reach a reasonable consensus
even when its members are, individually, reasonable.

Leigh