>
>No, it isn't - at least not mathematically. In math you use significant
>digits only - meaning, basically, only those digits that would affect the
>outcome in an operation. Since "20 + 1 = 21" and "020 + 1 = 21" the leading
>zero in the section operation is "insignificant" - it doesn't matter to the
>math. So it's not considered part of the number.
>
>
>
>This is a really simple explanation and a really bad explanation of
>"significant digit" which means something different in science. In more
>precise terms significant digits are only those that contain value, not
>those used just as place holders. In this way the number 60000, for
>example, has only one significant digit - the digit "6" - the three zeros,
>while important, don't contain any value and are just used to force the only
>significant digit into the proper place.
>
>
>
>Another, perhaps easier, way to consider this is the way numbers are written
>in scientific notation. When you write "60000" you've got a number, but
>information is missing as to purpose of the zeros - are they just there to
>push the 6 into place or were they measured? In this case you have to
>assume that they're insignificant.
>
>
>
>But if we write 6*10^4 then that shows us that 6 is the only significant
>digit. But you can also write 6.000*10^4 then that indicates that three of
>the zeros are indeed significant (usually that they were specifically
>measured) - however it also means that one is not. This happens a lot with
>measurements.
>
>
>
>Let's say I have a radar gun that could measure speed in 10 mph increments.
>When I clock a car doing "50mph" I know that the speed is at least 50mph but
>not more than 60mph. So the number "50" in this case has only significant
>digit: the 5 (since the zero is needed to "push" the 5 over, but was not
>measured itself). So in scientific notation you'd write "5*10^1" (well, I
>think - my scientific notation is rusty).
>
>
>
>But now a friend comes along and his radar gun measures in 1mph increments -
>when he clock's a car doing "50mph" he knows that the speed is at least
>50mph bot not more than 51mph. In this case the number "50" has two
>significant digits (since both digits were measured). You'd write
>"5.0*10^1" is scientific notation to indicate that the zero is indeed needed
>and significant.
>
>
>
>Now one of the most fundamental rules of determining significance (so
>fundamental that it's often not even mentioned) is that any zeros to the
>left of the last significant digit or decimal point are insignificant. They
>are doubly damned since they 1) don't change the value of the number and 2)
>can't EVER be measured (how would you specifically measure the zero in
>"021"?)
>
>
>
>CF is doing the right thing here since val() is a numeric function.
>
>
>
>Jim Davis
[Todays Threads] [This Message] [Subscription] [Fast Unsubscribe] [User Settings] [Donations and Support]
