On Thu, 13 Nov 2003 14:01:16 -0500, "John Gilmer"
wrote:



We don't have a perfectly round earth. That's the problem.

The meter, exactly like the nautical mile, and unlike feet and statute
miles, *is* based on the Earth.

WAS and not "Is."

Still "is" in the same sense as the nautical mile is.

You are right that the meter is no longer precisely defined based on
the Earth, and never really was, because nobody had an accurate
estimate of that before the first standard meter bar was constructed
in the 1790s.

But the same is true of the nautical mile, now most often defined as
1852 m. Just making sure that you aren't one of those who want to
have it both ways, claiming that the nautical mile is based on the
Earth and the meter is not.

Those "rational" folks who first invented the meter f*cked up.

No, they got damn close, when they took on a difficult task. We can
do a little bit better now, a couple of centuries later--but not much
better at all just using traditional surveying methods.

Measure an arc at sea level on Earth in grads, and it will still be
over 100 km in some places and less than 100 km in other places. In
other words, the meter as now defined still fits the earth. There is
no real significance in the particular meridian quadrant through
Paris.

Exactly how far off were they, anyway? If you go by the length of a
meridian quadrant on the Geodetic Reference System 1967 ellipsoid,
they'd be off by about 2.0 km in 10000 km. That's 0.020%, pretty
good. But if you go by the length of a meridian quadrant in the
WGS-84 ellipsoid, they are only off by 1.7 km in 10000 km--or is it
1.97 km? Whatever, it certainly differs from the value for other
ellipsoids that have been used, if you get down to the nearest meter.

How long is the actual meridian quadrant through Paris, if it were
measured as accurately as possible today? We don't know, because
measuring it is still difficult enough that nobody had even tried to
recalculate it. That 0.02% is a good ballpark estimate.

Still fits the earth well enough to that 0.01 grad along the equator
is about 1001.9 m, but 0.01 grad (geodetic latitude) going N-S across
the equator is about 995 m.

The meter
was re-defined as the distance between two marks on a Pt. rod kept at a
certain temperature. As technology marched on more "universal" standards
of length (and time, etc.) were invented. Usually the "new" standard was
close enough to the old for the accuracy required for most commercial,
surveying, or navigation purposes.

There is NOTHING magin about the metric system except for the fact that
units differ by powers of TEN.

There are several more important factors, including the fact that it
is used all around the world, including in many applications in the
United States.

The number 10 isn't even the most important number in the modern
metric system. The number 1000, with preferred prefixes those which
are powers of 1000, is also pretty important. But the number which is
really important is the number ONE. The SI is a "coherent" system of
units, as that term is used in metrology. That means that all the
derived units are some unitary combination of the base units.

This just isn't much of an advantage for most human activities. The "step"
size of 10 times is just TOO BIG!

A step size of 10 is 100 times too small. We'd all be better off if
the whole world forgot "centi" and almost nobody uses "deka-" and
"deci-" and "hecto-" anyway. But a whole lot a classroom time is
wasted teaching those useless prefixes. The CGPM ought to have had
enough sense to consign them to the same fate as "myria-" and all the
double prefixes which used to be acceptable, such as micromicrofarads
and hektokilometers and micromillimeters (aka millimicrons, where even
the micron name is no longer acceptable).


Gene Nygaard
http://ourworld.compuserve.com/homepages/Gene_Nygaard/