This entire article seems like some pretty obvious stuff that most people understand very well, wrapped in a veil of apparent profundity. First of all: it make

This entire article seems like some pretty obvious stuff that most people understand very well, wrapped in a veil of apparent profundity. First of all: it makes one very large assumption from the beginning that is never mentioned: "I, pg, can confidently identify a reasonable domain expert". That's basically a fallacy: this is the hardest thing to be sure of and Dunning Kruger applies heavily here (both for the apparent domain expert, and also for the person listening to their idea, who considers themselves such an excellent judge of the speaker's expertise). Beyond that, there's nothing profound here: any "average" person, when listening to a person they trust (and "trust" here means they believe that this person is competent, i.e. is a "reasonable domain expert"), will heed their ideas, and be more likely to consider the "crazy" ones. Paul is not unique nor advanced in this regard. The unstated problem is knowing whether you're good at identifying experts.