alt Hacker News> Because, apparently, a consistent trend of 100 consecutive falsifications implies nothing about the likelihood of #101. And at the end of the day, it's impossible to conclusively prove a negative.
That's right. Not sure why you sound a bit unhappy with this.
In particular, a source can become more untrustworthy over time if the source is repeatedly proven to lie or be reckless about the truth. I'm not sure you can apply the same logic to "categories of claims". What is the rationale behind your implied frustration that people are not "learning" that some "categories of claims" tend to be untrue? (not to mention the arbitrary grouping of totally disparate ones like Chem-Trails and Flat Earth)
If a “category of claims” has shared causal structure, then the category’s track record absolutely does tell you something about the next claim in it.
It’s not arbitrary. Alien UFOs, Chem-Trails, and Flat Earth are obviously all generated from the same distribution of bullshit: ambiguous or misunderstood phenomena explained by positing a vast hidden conspiracy.