Is the Scientific Method Valuable in Engineering?

Changing a variable hoping to notice changes _is_ forming an hypothesis contrary to what this article is claiming

For software engineering? It depends on how metaphorically you take "scientific method."

I mean, a program is....a big binary number. It is a mathematical object. It seems strange to be talking about investigating mathematical objects with the scientific method.

Sure, debugging is finding out something new about a mathematical object--but when we found out that the length of a diagonal to a square was irrational, that too was finding out something new about a mathematical object we already knew about.

Did we find out that sqrt(2) was irrational using the scientific method?

Debugging is like theorem proving--you are trying to prove the problems lies in certain part of the code, and then you are trying to prove your fix works.

Debugging is like a conversation between you and the original author of the code. You are trying to figure out what he said, and why it doesn't mean what he thought it meant.

Debugging is like a police invesgtigation at a crime scene.

etc etc. There are all sorts of metaphors we can use to describe debugging, but debugging is an utterly new thing. It was "discovered" by Maurice Wilkes, who said, " the realization came over me with full force that a good part of the remainder of my life was going to be spent in finding errors in my own programs."