“A few outsiders and weirdos saw the giant lie at the heart of the economy by doing something the others never thought to do: They looked.”
2.2 Do the numbers matter
As a premise for checking
2.3 Think stupider
Sometimes the simplest explanation is the right one
2.4 Repeated and overlooked information
Trustworthiness assessment relies on information that is repeated or overlooked
2.5 Boring, complicated, and consequential
The worst trustworthiness issues occur at the intersection of what is boring, complicated, and consequential
2.6 Is it just me?
Dealing with uncertainty about whether an issue is real
2.7 Avoiding playing detective
The burden of proof is on the original authors to substantiate their claims; trustworthiness assessment only has to quantify if there are issues in published work, not fully speculate or account for how they might have arisen. Speculation about intent is often actively unhelpful.
2.8 Highlighting issues does not require solving them
Not rewarding motte and baileys / evaluating articles based on their strongest claims
Authors should not be rewarded for publishing a paper saying “1+1=3” as long as they have a footnote saying “jk lmao”; the footnote doesn’t insulate you from critique, especially if it’s being cited for the claim and not the footnote