ahoward <ahoward / fsl.noaa.gov> wrote:
 may act differently since object would now _always_ be defined.  it seems like
> the 'new' rule will cause people to start writing
> 
>  loop do
>    object = 42
>  end
> 
>  object = nil
> 
> to undef objects things.  particularly with common var names like i, x, etc.

Why, though? I can't think of a case where I'd need to know a variable
was undefined, other than immediately following an assignment (and then
I'd want to know if the *object* I just pointed the variable at was
nil). In fact, the old behaviour always bothered me, since I couldn't
think of any reason I'd want the variable to go out of scope.

martin