msb@sq.sq.com (Mark Brader) (01/22/90)
There has recently been discussion in comp.lang.postscript of the North American typographers' units of measurement--the point, which is generally defined as 1/72.27 inch, and the pica, which is 12 points. At least two good-looking explanations of that curious number 72.27 were posted. This is one: > ... It is believed that, even though Fournier's pica was based on [i.e. > was 1/6 of] an inch, four thousandths were lost as a result of active > (and less than ideal) use of the original molds, and to the reproduction > process, as new molds were made to replace that which had worn out. [The "four thousandths" refers to 1 - 72/72.27, which is .00374-] And this is the other: | The 1886 standard from the United States Type Founders' Association was | based on the observation that the American-pica had an almost-relationship | to the cm: 83 picas ~= 35 cm. And so they standardized on exactly | "83 picas = 35 cm". There is, of course, no inherent reason why they could not both be true. However, nowadays the inch is defined as 2.54 cm. And 2.54*83*12/35 (where the 12, of course, is the number of points in a pica) IS NOT 72.27; it's a bit over 72.28. To have 72.27 points in an inch, 12 points in a pica, and 83 picas in 35 cm, the inch would have to be only about 2.5396 cm. This is where the cross-posting to sci.physics and sci.misc comes in. Can someone comment on just how much the lengths defined for the inch and the (centi)meter have varied in the past 104 years? I'd be surprised if either had changed as much as the 1 part in 6500 or so that's indicated here. Certainly there have been variations, but my impression was that they were on a much smaller scale than that. If I'm right, then either the 83/35 story is bogus, or there has been a further change in the point since then. (I'm rather reminded me of the story that the track gauge of the Toronto streetcar and subway systems, 58 7/8 inches, was chosen because it's exactly 1.5 meters. 1.5 meters is in fact a little over 59 inches. Yet this, too, *could* still be true, because gauges can be slightly adjusted after the system are in place...) Followups are directed to comp.lang.postscript; adjust if appropriate. Mark Brader "It is impractical for the standard to attempt to SoftQuad Inc., Toronto constrain the behavior of code that does not obey utzoo!sq!msb, msb@sq.com the constraints of the standard." -- Doug Gwyn