From: Pat Naughtin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
To: "U.S. Metric Association" <[email protected]>
Subject: [USMA:37818] Re: Centimeters
Date: Sun, 21 Jan 2007 18:25:55 +1100
Dear Tom,
Thank you for your detailed commentary on the use of the centimetre.
I have interspersed some remarks in black.
On 3/01/07 1:20 AM, "Tom Wade" <[EMAIL PROTECTED]> wrote:
>> It seems to me that applying the adjectives 'perfectly good' to the
>> centimetre following such observations is inappropriate.
My objection to the value judgements implied in the words, 'perfectly' and
'good' are simply based on my observations that metrication transition
processes are much, much, slower when centimetres are used rather that
millimetres.
Let me clarify the basis of my personal observations. I have been observing
metrication processes and attempted metric conversions closely since 1970
and I have yet to see any fast, smooth, or economical metric transitions
that involved using centimetres. Typically, the use of the centimetre
simply
leads to a long, protracted, rough, and expensive process where metric
conversions dominate both TO metric units but more often FROM metric units
back to the old pre-metric measures.
However, having said that, I know the status of the centimetre both
historically and in the current BIPM brochure.
> The centimeter is certainly a 'perfectly good' unit. When it comes to
> standardizing within an industry, the millimeter is far more likely to
> be the more appropriate unit, but this is not necessarily always the
case.
My assessment based on Australian job titles is that millimetres are used
in
about 84 % of trades, crafts, and professions; centimetres and inches are
still used in about 10 %; and the rest such as butchers (6 %) don't measure
much length anyway.
> In any engineering discipline, the mm will be preferable, as it gives
> greater precision without the need to resort to decimal places. It does
> end up with more digits on large measurements than the cm does, but the
> greater precision more than compensates for this. If I am doing any DIY
> work, I always measure in mm.
To the best of my knowledge, the issue of more digits on large measurements
has never proved to be a problem in any of the trades in Australia. The
advantage you mention of not having the need to resort to decimal places is
further enhanced by not having to refer to fractions at all not common or
vulgar fractions and no decimal fractions. There is only pure number that
can be readily manipulated with any cheap calculator.
> The comparison between the building industry of Canada and Australia
> (the former adopting cm and the latter mm) is illuminating. Australia's
> transition has been more rapid and complete. Even allowing for Canada's
> proximity to a much larger country still using Colonial measure, it is
> still reasonable to conclude that choosing the mm was a better decision
> *for that industry*. It is *not* enough to conclude that the mm is
> better for all industries.
I agree that the comparison between Australia and Canada is valid but I am
on shaky ground in so far as my own personal knowledge of Canadian
metrication is concerned. A lot of my conclusions are based on extensive
correspondence with the late Joe Reid, President Emeritus of the Canadian
Metric Association.
> Note that for an industry that is converting, there will be a natural
> internia against moving away from something familiar. The new units
> must offer a significant advantage in order to overcome this. Using mm
> meant moving from awkward constructs such as 3' 5 3/8 " to a whole
> number (1034 mm). Where cm are used the conversion is 103.4 cm. The
> second is less attractive because you still need the decimal point (a cm
> is too coarse for this type of design to be able to round it off). Thus
> the perceived advantage in going to metric will be less when cm is used.
It's probably worse than you say. Many people when they see 103.4 on their
centimetre tape round this to 103 1/2 because they are used to halves,
thirds, quarters, and even eighths of yards, feet, and inches. I have seen
this approach used extensively in the textile industry where I worked for
about 13 years. Of course, carrying you fractional skills and your mixed
number skills (as in could I have 2 metres and 30 centimetres please) is,
in
my view, inappropriate and unnecessary.
> Comparing the textile and building industries in Australia, and assuming
> that because the former adopted cm and is slower to migrate means that
> mm was more appropriate has a subtle flaw. The flaw is the assumption
> that if the textile industry had chosen mm, then it would have been
> faster. To be able to conclude this, you would need to compare two
> textfile industries, one of which converted to cm and the other to mm.
> I find no example of this cited in the extensive research.
I totally agree with you. Like you, despite extensive searching, I have not
been able to find a single example of any part of the textile industry who
has not chosen the centimetre pathway. However, that said, I have also
observed that all of these have, since their centimetre decision,
experienced slow, protracted, difficult, and extremely expensive metric
conversions most of which have yet to be successful even after some 35 to
40
years. I don't think that this is a coincidence, but to concede your point,
I suppose that it could be.
> The truth is, for clothing sizes, the cm is the more appropriate unit to
> use. Pre-metric sizes used a resolution of half an inch, which is very
> close to 1 cm.
Only for some things is a tolerance of 10 millimetres needed. Shirt collars
are one example. Even there, bespoke tailors eschew measurements and cut a
scrap piece of fabric to fit their client's neck. Neck comfort is crucial
to
the performance of a shirt and I prefer personally better fit than 10
millimetres. I have been know to move the top button on a troublesome shirt
by about 5 millimetres.
> You don't need mm precision for meauring trouser size.
> If my waist size is 852 mm, and another person's 847 mm, we will both
> take a 85 cm trouser size.
If your waist size is 852 mm and you live in Ireland where the centimetre
was chosen as the small measure, your trousers will probably still be
offered to you in inch sizes loosely disguised by 'nominal centimetres'.
It works this way. The trouser maker will decided that his 38 inch waist
trousers are his main seller and he will go up and down from there in lots
of 2 inches. To hide the fact that he is not using metric he will nominally
call the median size '97 cm' and then proceed up and down from there in
lots
of 5 centimetres. His trouser rack will contain these sizes:
82 87 92 97 102 107 112 117 122
So, using your examples of my waist size is 852 mm, and another person's
847
mm you will be given the choice between a size 82 and a size 87. Note that
these metric conversions have now subtly become size numbers and not
measurements that could be used in a court of law for ill fitting garments.
Young shop assistants soon learn that they have to learn about the real
measures in inches behind the 'nominal centimetres' so that they can be
accepted into the community of men's outfitters. I expect that this policy
should last for another 100 or more years.
> Being able to group people into standard
> sizes is very important in reducing manufacturing costs and shop stock
> size.
Agreed, and that's why they used an interval of 2 inches in the men's
trouser trade and they seem to want this to continue as a rite of passage
for trainee shop assistants. It seems to me to be a power thing rather than
an issue of measurement.
As a side issue, the Australian women's clothing trade opted for
centimetres
with a 5 centimetre interval between sizes but they decided to start at a
hip size of 95 centimetres and then to go up and down from there with these
sizes numbers:
80 85 90 95 100 105 110 115 120
Note that these are actual sizes and not the 'nominal centimetres' that are
hidden inch sizes in men's garments.
> Trying to use mm means either having a much greater range of sizes
> (which impacts costs) or tagging an unnecessary 0 on to every measure.
This is not necessarily true. You could still choose to produce a range of
men's trousers using intervals of 50 millimetres.
The tagging an unnecessary 0 on to every measure would in fact be an
advantage to Australian men who work in the 84 % of Australian jobs who use
millimetres all day every day. You don't have to try and get your head
around the centimetres that you haven't used, or even seen since the day
you
left school, when you go into a clothing store. You just use the
millimetres
you have been using all day.
> The reason why a transition to metric will not be so fast here is that
>
> (A) the preceived advantage is not as great: moving from half-inch to
> whole cm is nice, but not as attractive as moving from awful fractions
> of inches combined with feet to whole numbers.
I can see little benefit in a transition from inches, feet, yards, and
their
attendant halves, thirds, quarters, and eighths, to a regime of millimetres
(for buttons for example), centimetres, and metres that might all be
divided
into not only halves, thirds, quarters, eighths, and sixteenths, but also
into tenths, hundredths, and perhaps other decimals.
> (B) body sizes are some of the most difficult things for people to shift
> units about.
I don't agree that 'some of the most difficult things' because if they are
badly planned all metric transitions can seem to be difficult. Recall that
France tried a voluntary metric transition in 1795 this proved to be too
difficult; France then tried a metric conversion by redefining the aune,
the
boisseau, the livre, and the toise in metric terms this proved too
difficult; and then finally, after almost 50 years they decided to try
direct metrication which was successful within a year or so and they
haven't looked back since. In my opinion, it is the process that you choose
for your metrication transition that is most important in providing for a
smooth or difficult metric transition and that is based on the choices
you
make when you begin planning your metric transition program.
> The real crunch is would the transition have been faster if mm were
> chosen. I don't think so. In fact given the unncessary precision
> implied, I think it would hinder it (in the same way that people get
> turned off metric when they see ridiculously precise conversions).
> Whatever chance you have of hoping people will get used to replacing 34
> inches by 85 cm there is no chance of 850 mm being accepted.
An interesting refutation of what you say happened in the leather and shoe
trade. They have been beset with measurement troubles since the Magna Carta
in 1215, and the subsequent Assizes in 1304, decreed that the barley corn
would be the standard of length for all of England. The shoe makers chose
the length of a barleycorn as their standard 'size' but they did not yet
specify where 'size 0' might be.
Because leather is so expensive, modern shoe designers measure in tenths of
millimetres where a shoe might have an overall length of 2800 tenths of
millimetres (280 mm) that might be divided as 1000 from the heel to the
tongue, 1100 from the tongue to the start of the toe case, and 700 for the
toe case. This shoe would then be labelled with a range of 'Size' numbers
based on the lengths of barley corns. By doing this the shoe maker can aim
for tenths of millimetres accuracy but still leave the actual choice as to
whether the shoe fits or not to the customer's profound knowledge of barley
corns in the 13th and early 14th centuries. Notice that the somewhat odd
choice of tenths of millimetres has allowed the shoe factory workers to
avoid fractions altogether.
> Measuring people's height is also something in which mm precision would
> be daft. Of course, you can try to split hairs by saying that 174 cm
> should be quoted as 1.74 m, thus using meters rather than cm, but you
> are still measuring to cm accuracy, so this is really 'hidden
> centimeters'. If you are arguing that one of the most powerful
> advantages of mm is that it gives you whole integral values (and I
> agree) then the same argument applies that 174 cm is easier than 1.74 m.
> For reference, I always write my height as "1.74 m" but speak it as "one
> seventy four".
I, too, have had trouble with this issue until I realised we could consider
our old estimating abilities in feet and inches. I don't know the
experience
in Ireland, England, or the USA, but here in Australia, it was usual for
people to guess the heights of others in ever numbers and in lots of 2
inches and this translates to 50 millimetres or preferably, in my opinion,
0.05 metres.
I once spent some time training police officers and the estimation of
heights was a critical daily issue for them. They were used to guessing in
lots of 2 inches, viz: 5'2". 5'4", 5'6", 5'8", 5'10", 6'0", 6'2", 6'4",
6'6", and so on. I tried unsuccessfully to teach them about centimetres and
they absorbed this in the same way as the tailors. They took 6'0" to be 183
centimetres and then moved up and down from there. Their series went like
this:
158 163 168 173 178 183 188 193 198
Clearly this encouraged young police officers to learn about inches from
their senior colleagues because these were the true underlying measures.
I then chose to use metres in the form of decimal numbers with two decimal
places that were to be ended in either a '5' or a '0' and trained them to
use standard doorways (2.05 metres) and their own heights (generally huge
they were big men) to guess the heights of the people around them. This
technique proved to be successful and with a little training they became
quite accurate. I suppose that I could have persisted with centimetres but
I
didn't and the added benefit was that these officers became quite
interested
in calculating their Body mass Index using their mass in kilograms and
their
height in metres.
> I would certainly agree that in most industries, mm should be chosen
> rather than cm as the increased precision and whole numbering is
> decisive. But this doesn't mean it is *always* appropriate, and it
> certainly should not mean that the cm should be 'deprecated',
> particularly in everyday use. The cm is a very handy size for
> approximate measure, and the last thing you should do when you are
> trying to get people to convert to metric in their everyday life is to
> try and force them to use something less convenient.
But this is my exact point. The centimetre is not more convenient because
it
(always?) brings with it the old baggage of fractions. Consider the problem
of the software engineers at Microsoft when they wrote Microsoft Word. If
you try to change your default rulers to metric you have two choices:
centimetres or millimetres, where the centimetres are divided into quarters
and the millimetres are arranged in lots of 2 1/2.
The other issue that centimetres bring in a metric transition is metric
conversion. People seem to think that it is appropriate to convert from
centimetres back to inches. This is time consuming and enormously
inefficient. In a large textile mill it is interesting to watch the
interactions between the machine operators and the technicians who repair
and set up the machinery. The latter use millimetres exclusively while the
former use a combination of centimetres and inches it leads to some
strange conversations.
> Lastly, the only metric quantity that all Americans understand and are
> completely familiar with is money. Everyone knows that $1.84 and 184 c
> amount to the same value. This parallel with meters and centimeters is
> too good a comparison to miss when trying to convince people that the
> metric system is actually easier than what they are using (explain
> pounds, shillings and pence to them, and they will agree this was an
> awful cumbersome system; then point out that they are using the
> equivalent in measuring everything else).
What you say is true, but you and I both know that there is another reality
emerging from behind the closed doors of metrologists, engineers,
architects, and other professional metric unit users, and this is the
preference for metric prefixes that are represent in groups of three
decimal
places. This use of preferred prefixes has already, more or less, killed
off
deci, deca, and hecto, and now centi is the only one of the original metric
system prefixes left. Let me repeat, I know the status of the centimetre
both historically and in the current BIPM brochure so their demise is due
to
movements in the perception of practical value rather than in their formal
acceptance by metrology officials.
> So please lay off the centimeter. Just because it is not as appropriate
> to adopt in the case of the clear majority of industries does not mean
> it is not a useful, widely accepted and 'perfectly good' metric unit.
I know that I stand alone on this issue, but I am also aware that if anyone
who is about to choose a metrication pathway chooses to use centimetres
they
are in for a difficult, slow, and expensive transition to the metric system
and my only goal in talking about the centimetre at all is to try and help
others avoid that fate. I've been there in the textile industry and it was
not a happy experience. I, for one, will not be using the value laden
adjectives 'perfectly good' when I am discussing the centimetre.
Cheers and thanks again for the opportunity to explore this important issue
further,
Pat Naughtin
PO Box 305, Belmont, 3216
Geelong, Australia
Phone 61 3 5241 2008
Pat Naughtin is the editor of the free online monthly newsletter,
'Metrication matters'.
You can subscribe at http://www.metricationmatters.com/newsletter
Pat is also recognised as a Lifetime Certified Advanced Metrication
Specialist (LCAMS) with the United States Metric Association. He is also
editor of the 'Numbers and measurement' section of the Australian
Government
Publishing Service 'Style manual for writers, editors and printers'. He
is
a Member of the National Speakers Association of Australia and the
International Federation for Professional Speakers. See:
http://www.metricationmatters.com
This email and its attachments are for the sole use of the addressee and
may
contain information that is confidential and/or legally privileged. This
email and its attachments are subject to copyright and should not be partly
or wholly reproduced without the consent of the copyright owner. Any
unauthorised use of disclosure of this email or its attachments is
prohibited. If you receive this email in error, please immediately delete
it
from your system and notify the sender by return email.
