Wasn't it Dave Dunn who wrote:

>At the risk of flogging this topic beyond death, I would like to point out that
>the default values for translate and rotate are <0,0,0>, but that the default
>values for scale are <1,1,1>. In other words, and object that has not been
>translated has a default translation of zero on all axes because it hasn't been
>translated yet. The same object, if it has been created at a given size, is
>scaled to 1 on all axes. In this way, we can move an object along x by typing
>"translate x*4," because this is, in effect, a translation of <4,0,0>. In order
>to maintain consistency at the user level (I keep hearing that things should
>hold true in every case, a scale of our object along y by 2 should be able to
>be written "scale y*2," returning the defaults, for an equivalent expression of
><1,2,1>. IMO, the warning (and certainly not an error) should not be given,
>because the original assumption that default scaling should be <0,0,0>, is
>wrong.

So, what you're saying, is that when the code that implements "scale" is
passed a value <0,2,0> it would have to look back at the source code and
fine where those values came from. If you had explicitly specified
"scale <0,2,0>" it should display a warning, but if you had originally
specified "scale y*2" the warning should be suppressed. This would get
really tricky if you had written "scale My_Vector" or "scale
My_Macro(2)".

-- 
Mike Williams
Gentleman of Leisure

Post a reply to this message