On 1 Nov 2000, at 11:04, Hooper, Bill and or Barbara wrote:
> Many of you have countered my argument in favor of voluntarily phasing
> out the use of the special prefixes centi, deci, hecto and deka. The
> arguments always seem to center on the centimetre.
> But let me draw a comparison between length and mass:
> LENGTH
> If we wish to express the length of 152 cm without using the centi (or
> deci) prefixes, we could use either 1.52 m or 1520 mm. Several of you
> have argued that 1520 mm is too long and furthermore implies a
> precision of 1 mm which may not be warranted.
Ok, but please change the size ... (one could argue that 5 ft is
simpler) .
Precision is important. If you say 1.52 m is ok, but suppose the
size is 1.50 m ( that is 150 cm : some one could have the
tempation to say 1.5 m ! ) ... so using the cm is automatically
implying that you have a 1 cm precision ( or say 1.50 m +-0.005 )
> But you miss the point when you assume I am arguing in favor of using
> 1520 mm. I am not. I am arguing that one may use either metres OR
> millimetres, WHICHEVER IS MORE SUITABLE. If 1520 mm is not suitable
> (as noted above), then use 1.52 m. Where's the problem?
None, just an habit: multiples/submultiples are allowen, you are not
forced to use them.
> MASS
> Now please compare this with a similar situation involving mass.
> Suppose something has a mass of 1.52 kg which is also 1520 g. Does
> anyone seriously think it would be a good idea to express this in
> centigrams: 152 cg? That would be perfectly valid SI usage but I think
IT IS NOT A LEGAL USAGE: due the unfortunate choose of base
unit 152 cg are 0,00152 kg .
1.52 are 152 dag .
> it would be highly inappropriate. No one uses centigrams for measuring
> anything. Let's not start using centi there, where it is not already
> in common use.
HONESTLY, i have never seen non thousand submultiples of g in
common use, neither the dag .
howewer i have seen in use, as i told some time ago the dat , kt
,Mt Gt and Tt , much more frequently than them Mg Gg Tg Pg
(and remember that there is no equivalent for the dat, that is a
reasonable size (in the use i have seen you have always integer
size, with the right level of precision)
> If we don't need centi for centigrams, then I fail to see why we
> really need centi for centimetres. The situations are the same.
> (Notice I wrote "NEED", not "WANT".)
Do you need the entire line of multiple/submultiples from Z to z for
every unit ?
The fact that one multiple is not used for one unit, is no reason to
discard for others
> We are unfortunately stuck with the historical fact that centi is in
> VERY common use for lengths in centimetres. Fine! So we can accept the
> continuing use of it there. But let's not start measuring masses in
> centigrams (and time in hectoseconds or voltage in dekavolts or power
> in deciwatts, etc).
Please, no, at least in your personal cg !!!!
> The base unit plus the prefixes kilo and milli will suffice for
> virtually ALL common measurements (and the other powers-of-1000
> prefixes will suffice in any case where killo and milli are not large
> or small enough).
> PS I appreciate this vigorous exchange of ideas. I hope no one
> interprets this discussion of our varying views to indicate that we
> are not all united in promoting the general adoption of SI.
Leonardo Boselli
nucleo informatico e telematico
Dipartimento Ingegneria Civile
Universita` di Firenze
V. S. Marta 3 - I-50139 Firenze
tel +39()0554796431 fax +39()055495333
http://www.dicea.unifi.it/~leo
