> From: Gustaf Sj�berg <[EMAIL PROTECTED]>
> Subject: Re: Cribsheet
>
> Since the fractions around the baseunit are the most used, they deserve there
> own names.
I will agree with Gustaf that this is a reasonable argument in favor of
using the minor prefixes (deci, centi, deka and hecto). But it is just one
reason and there are equally good reasons not to use them.
> Then there is the accuracy. Should you measure everything exactly to the
> millimeter all the time? The accuracy of +-1mm or +-1mL can't be
> kept in all situations. That is totally unecessary.
Accuracy is not an issue. the use of a unit does not always mean that the
quantity has been measured to plus-or-minus one of that unit. For example,
we commonly refer to the distance from Earth to the sun
as 93 000 000 miles in olde English units
and 150 000 000 kilometres in SI. But no one presumes that these numbers
indicate that the distance is being given to a precision of 1 mi or 1 km.
Are we closer to the sun in the mile high city of Denver, Colorado so that
the distance to the sun must be given
as 92 999 999 miles, and are we closer in the kilometre high city of
Ashville, North Carolina so that it
is 149 999 999 km? Of course not!
If we DO choose to make our numbers indicate the appropriate precision, then
in SI we could call the Eart-sun distance 150 Gm (gigametres) which would
imply a precision of 1 Gm which is one million kilometres, and probably is
closer to the truth. I note with smug satisfaction that the ability to do
this does not exist in the olde English mix of units.
Actually, writing the distance in terrametres is probably better -- if the
precision of the value that is one of the primary concerns -- because then
it can be written as 0.150 Tm if it is is precise to about 0.001 Tm or can
be written as 0.15 Tm if it is only precise to about 0.01 Tm.
One more example (bear with me, please). My height is 181 cm. If I choose
not to use the centi prefix I can report it as 1810 mm. If I fear someone
will think that means a precision of plu-or-minus 1 mm (so my height must be
between 1809 mm and 1811 mm), then I can report it as 1.81 m and the
"problem is solved. Frankly, I would never fear that someone would
misinterpret a persons height of 1810 mm that way.
regards,
Bill
