Martin B?hr wrote:
>On Thu, Jan 15, 2009 at 12:12:41PM +0100, Stephen R. van den Berg wrote:
>> This behaviour is perfect. It is exactly what I would expect, and therefore
>> there would be zero benefit to me to change that behaviour.
>what is perfect about undefined behaviour?
Undefined behaviour is "perfect" because it means that the implementation
can choose the most efficient algorithm available, without regard to any
special treatment I don't expect anyway.
--
Sincerely,
Stephen R. van den Berg.
The ultimate Darwin Award winner:
Going first in a Russian roulette game with a normal gun (no revolver).
|