Comments interspersed. >I see a lot of discussion about non-SI metric units, and I'm not >sure I agree >with the general message that non-SI units (like centimeters or hectares) >should never be used. I'm challenging you purists to convince me!
Where do you get the idea that centimeters are non-SI? Pat Naughtin has pointed out that the Australian building trades have had success with using millimeters exclusively for the various components, including plywood, other sheet material, and lumber. However, that is an industry practice, not a hard and fast rule. Meter (m), with any valid prefix, is an SI unit. In practice, people don't find decimeters (dm) and decameters to be very useful, but they are valid nonetheless. > >I think the non-SI units that aren't based on factors of 10 of the >base units >should be completely eliminated, including > Torr & millimeters of mercury > horse power (defined in terms of kilograms of force) > calories and Calories I don't think anyone would disagree with you on that. >Also, for scientific work, I think we should all use mks units >(which is SI) >and not cgs units (which are not SI). So I would dump derived units like >ergs and dynes. The history is cgs -> mks -> mksA -> SI. The first three are all obsolete, although SI is very close to mksA. > >However, in daily use, I think that units like cm can be handy. >The reason is >that most normal humans want to use a nice round integer that is >as small as >possible without losing required accuracy. Decimals are somewhat >clumsy for casual use. > >For starters, the SI unit of kPa is not as well suited for >barometric pressure >as the millibar. It feels a lot more "human" to say 1014 millibars than >101.4 kPa. If you want to be "human," use the approximation -- 101 kPa. On the other hand, if you prefer the scaling of millibars, use hectopascals (hPa). Many people object to hectopascals, but they are correct SI. > >A second example is height. Precision at the millimeter level is >completely >unwarranted. (Measure your height when you first wake up and then before >going to bed and see for yourself how meaningless that last digit >is!) So I >think it's simpler and more convenient to say someone is 181 cm tall than >1810 mm. I haven't seen anyone here suggesting the use of millimeters for our height (unless I've missed something). Centimeters are the customary units (on SI-compliant forms, licenses, passports, etc.). > >Beverages are a third example: 2 dL, 3 dL, 4 dL, or 5 dL. Nice, >clean simple >integers give a good range of beverage sizes from small to large. Customs vary. In the U.S. and Canada, milliliters are normal for standard wine bottle sizes and the smaller liquor bottles (e.g., 750 mL). (In France, centiliters are more common.) For the larger wine bottles (e.g., 1.5 L) and liquor bottles (e.g., 1 L, 1.75 L), liters are customary (after all, 1000 mL seems a little silly). I'm happy to see 200 mL, 400 mL, etc. Deciliters would be nice for those of us who work with all the prefixes. However, SI will gain acceptance with fewer, rather than more prefixes in everyday use. Because of the way wine and liquor are packaged, the public is reasonably comfortable with liters and milliliters. Large soft-drink bottles reinforce the familiarity with liters. Shampoo and other household products (especially from Procter and Gamble) reinforce familiarity with milliliters. > >Farm size is a fourth example. Farms tend to be a few hundred hectares in >size. It's convenient to buy, sell, and price land in hectares. >I think you >would be crazy to suggest farmers measure, buy, and sell land area with 7 >digit numbers of meters squared. How much should a square meter >of farm land I like hectare, myself. However, it's a multiple of the are, which is not an SI unit, which is why its use is deprecated in the SI Brochure. If 'are' were to be given the same special status as liter (non-SI unit approved for use with SI, avoiding the rather long "cubic decimeter"), it might be helpful. The only trouble, there, is that nobody actually uses the are as a unit (except as prefixed, in hectare). >cost? Metric prefixes don't work with square meters. Is a "mega meter >squared" the size of a small farm or larger than Texas? Finally, >note that >"square kilometer" is NOT and SI unit. Of course they do. And square kilometer (km^2) IS an SI unit. The exponent applies to the whole unit, as prefixed, so 1 square kilometer is 1 million square meters (1 km^2 = 10^6 m^2). I suggest you read the SI Brochure (http://www.bipm.fr/pdf/si-brochure.pdf) if you want your knowledge of SI to be thorough. Bill Potts, CMS Roseville, CA http://metric1.org [SI Navigator]
