The reason I explain it like I do is that "observation" and "theory" are the core elements; everything else (all the notes and gotchas) are constraints, require

The reason I explain it like I do is that "observation" and "theory" are the core elements; everything else (all the notes and gotchas) are constraints, requirements, or clarifications of those two elements. For example, falsifiability is a constraint on what qualifies as an allowable scientific theory. Recognizing that theories are probabilistic leads to the understanding that they remain provisional (with probability never quite reaching 1). Testing a hypothesis means making observations of a particular sort and comparing them to specific predictions from the hypothesis. Making data and algorithms public is a way to ensure observations can be passed from person to person with a minimum of bias creeping in. And on and on and on...