When we use a single word for a complex phenomenon (in this instance, the "scientific fact") we bound to have endless discussions talking past each other. The

When we use a single word for a complex phenomenon (in this instance, the "scientific fact") we bound to have endless discussions talking past each other. The focus on "95%" gives away that the focus of the article is on a preliminary stage of scientific inquiry, where uncertainties around the measurement (observation) process are dominant and the explanatory framework of what is happening is sketchy (statistical) in nature. This is typically the case when the system we want to establish scientific facts about is very complex (e.g. biological or much worse, human social). Inceased scientific confidence obtains only when we can control that complexity and zoom-in into more fundamental behavior that can be both quantified and explained in more detail. That deepening of understanding leads to a cycle of prediction/measurement/experiment/confirmation that keeps adding "confidence" in a way that has nothing to do with statistical "95%" confidence. E.g., Netwonian laws were "proven" to very high certainty and yet they got eventually replaced by even more accurate scientific models. Scientific facts are a journey, not a destination. For a sane discussion we need to clarify what phase of the journey we are talking about.