> where does h go when you compute a derivative? After cancelling it from everywhere we can, we set it to zero. > Why can't you just set it to zero? Before c

> where does h go when you compute a derivative? After cancelling it from everywhere we can, we set it to zero. > Why can't you just set it to zero? Before cancelling it as much as we can? Because generally we most of the time would end up with something like 0/0.