Its not obvious at all to me the result should be a float. I think the times I'm going to accidentally multiply a float by an int outweigh the times I want to lazily multiply a float by an int. And, some percentage of that time I would want an int anyway!
