What are we doing in science?
We are trying to find statements whose truth or falsity will be interesting, and we are trying to assign a corresponding probability, hopefully high in either direction. We call this probability to be a [what?]. To form one, we must "route" existing evidence towards the new statement.
Statements can be interesting for many reasons. Strong judgements about them may be considered difficult or unlikely, or maybe they are helpful as evidence to judge some other interesting statement. Statements can refer to judgements, or to the routing proceedure, or to effective truths such as "I observed Y after doing X". Statements like "I observed Y after doing X" if true are not interesting due to impressive routing of evidence, so they must be interesting by being unexpected. Building a fusion reactor is the only evidence needed that people can today build funsion reactors. Part of the work to be done is to find statements where you feel you have inside knowledge that makes your judgement of a statement very different than the widely accepted one. You are mining for judgements, alongside a million other miners. To suggest that the process of finding suitable statements to work with is itself interesting is to make this process also a subject of interesting statements. What are the flaws of current research directions? What new analogies have appeared that can do some inference routing work? What is a better pick axe, or why is the next valley more lucrative than the current?
Quote from E.T. Jaynes:
To form a judgement about the likely truth or falsity of any proposition, \( A \), the correct procedure is to calculate the probability that A is true:
\[ P(A | E_1, E_2, ...) \]conditional on all the available evidence, \( E_1, E_2, ... . \).