Re: [boost] Interest in dimension/unit-checking library?

From: Scott Schurr (scott_schurr_at_[hidden])
Date: 2005-10-13 13:19:18

I'm no where close to the expert that you guys are, so I'll quickly
toss in my two cents and get out of your way.

On 10/13/05 Matt Calabrese <rivorus_at_[hidden]> wrote:
>
> On 10/13/05, Deane Yang <deane_yang_at_[hidden]> wrote:
> >
> > Mathematically, I think Andy is entirely correct in what he says above.
> > Once you take the ratio of two measurements of the same dimension (or
> > quantity), you end up with a pure number that really can't be
> > distinguished from other ratios or pure numbers.
> 
> 
> I tend to disagree here. A ratio of two lengths, for example, is in 
> fact a value in radians. 

Um, careful there.  Suppose I'm measuring the height of a shrub at
various points in time.  If I take one of those measurements and 
divide it by a measurement taken at a different time, then I get a 
ratio that expresses the shrub's change in height in a unitless 
way - I get the same value whether I made my measurements in meters, 
inches, or furlongs.  This is an example of a common case where 
dividing two lengths doesn't give radians - it's just a ratio.

The dimension/unit-checking library should solve the problems that
it has enough information to solve.  When it runs out of information
it should ask for help.  Would it make sense to have something like
a unitless_cast<radians>?

Scott Schurr