I'd like to comment on this idea of the changing inch as it illustrates the problem of being too liberal in rounding a conversion factor. Measured data should be rounded consistent with its measurement accuracy. However, conversions are either declared values (legal definitions) or the highest accuracy a lab is capable of. Destroying that accuracy is frought with with risk. The UK had their own problem with the "incredible shrinking yard." However, according to NIST SP447 (downloadable from their site) p. 21, the US yard prototype was measured in 1893 just before the Mendenhall order. It was determined to be 0.914 399 80 m, only 0.22 parts per million different from the International foot adopted in 1959. Yet, the Mendenhall order established a rounded value 1 m = 39.37 inches (equivalent to 0.914 401 83 m) which led to a greater difference with the UK yard, and became so entrenched that we are still cursed with it today in the form of the US Survey foot. They knew it was wrong, but the chose to adopt a definition that had good enough accuracy for commercial purposes from the Metric Act of 1866. The Mendenhall order freed us from separate physical standards for Customary measure, which is a good thing. However, the wrong choice of value, and a worse choice than they were capable of making at the time has not worked out entirely well. In any case the difference between Survey and International foot is only 2 parts per million, affecting volumetric measure 6 ppm, so it hasn't changed much. I disagree that we should round conversion factors willy-nilly to a "nice" round number when we know better.
--- On Sun, 4/5/09, Jeremiah MacGregor <[email protected]> wrote: From: Jeremiah MacGregor <[email protected]> Subject: [USMA:44373] Re: Even with "dual," you can't please everybody To: "U.S. Metric Association" <[email protected]> Date: Sunday, April 5, 2009, 9:08 AM I interspersed some remarks in GREEN below. It's probably not an over exact metric conversion. It looks like it was converted from an over exact cubic inch. This is a very interesting point. Since the inch has change noticeably with time then all the volume measurements have also. I wonder when the NIST formulated their conversion factors and if they took into account the changing value of the inch. Maybe it is time for the NIST to officially redefine the spoons and cups to the metric sizes that the spoons and cups are made to and not the nonsense they want them to be.
