I read the article as against the word not being scientific enough because it doesn't differentiate between the competing theories. He may have a point for spec

I read the article as against the word not being scientific enough because it doesn't differentiate between the competing theories. He may have a point for specialists in the field, in specific circumstances, a layman may just want to communicate the concept without knowing the details. To use software terminology, he's arguing against abstraction in favour of implementation specifics.