Discuss the following:
(a) Why is the Normal distribution so important, particularly with respect to problems dealing with statistical inference?
(b) Ok, now that we accept/believe that the Normal distribution is so widely applicable, computationally why is it so useful?