I think as often as not, engineers hit on something that works, and later on the scientists come along to explain why it works, and figure out a framework of rules around the principle that allows it to be systematized.

I think that the "it works part" is science ( measure quantify qualify) this is what the software community does really badly.

> Yup, Engineers use science, craftsmen use intuition.

Sometimes. The early history of radio was one instance where the engineering ran ahead of the science.