Le lundi 01 décembre 2008 à 11:15 +0000, Mark Dickinson a écrit :
> My initial reaction to this was negative, but I'm struggling to think of 
> situations where it would be bad.

Consider someone who writes:

z = y / x
return my_list[z]

In all his tests, x is a divisor of y and therefore z is an integer, the
code runs ok.
Suddenly in an use case, x is not a divisor of y, z is therefore a
float, and the "return" line raises a TypeError.

The reverse can also happen, consider something like :

z = y / x
return z.as_integer_ratio()

As for :

<type 'int'>
>>> type(2**-3)
<type 'float'>

I'd argue it is less annoying because it only depends on the value of
the second operand which is, most of the time, a constant, and therefore
you know upfront if the result will be a float or an int.