Jim F. pointed out
> It was a loose figure of speach, Bill. Yep, one year is
> nearly pi times10 Ms in length.
That's an interesting fact and I have often used it. But, for what it's
worth, it is also true that a year is closer to 10 times the square root of
10 megaseconds. Using the following:
pi = 3.141 59
sq.rt.(10) = 3.162 28
and calculating the error by comparison with
1 year = 31.56 x 10^6 s
I calculate:
10 x pi Ms = 31.42 x 10^6 s (error of -0.44%)
10 x sq.rt.(10) Ms = 31.62 x 10^6 s (error of +0.19%)
The difference in the errors is pretty trivial, so either
10 x pi Ms or
10 x sq.rt.(10) Ms
is a useful approximation to one year. Pi is probably easier to use since
more people know what pi equals than know what the square root of ten is,
and it is fairly easy to say "ten pi million megaseconds". But the square
root of 10 does have the advantage of invoking another "metric" 10 into the
equation, and being just a little closer.
