Hacker News Reader

Comment by cousin_it | Hacker News Reader

Here's one possible solution: allow public figures to lie all they want, but also let them enter their advance predictions on some kind of permanent record (ironically, the blockchain would be pretty good for this). Whenever you meet a new pundit and don't know if they should be trusted, go check the chain. If they never registered any predictions, or it was all vague stuff, or specific but didn't come true, then just ignore them.

More hardcore variant: instead of predictions, keep a record of bets, or imaginary scrip in some kind of prediction market. Maybe call it "AccuracyCoin". If someone has a lot of it, that means they were right on a lot of bets where many others were wrong, so you should pay attention.

View on HN

The biggest problem is figuring out the truth from lies and mistakes and competing narratives. Maybe the pundit predicted that unemployment would go up, and some say it did, but others explain that it's not actually up because people are leaving the workforce voluntarily, others claim that it was miscalculated in the past, yet others claim that it's actually a bad measure that shouldn't be trusted at all.

There are very few directly testable predictions. Even something as basic as whether the Earth is flat is hard to conclusively prove without enough background - how could you hope to understand the economy?

I imagine predictions about unemployment would be based on numbers published by specific agencies. For example, see this Metaculus question: https://www.metaculus.com/questions/6296/us-unemployment-abo...

The question is, why should I believe those numebrs? "Unemployment" is a pretty technical measurement (the number of people who have not been employed for some amount of time X < T < Y but are still looking for a job) that is estimated through a pretty complex process (surveys and all of the statistical analysis that has to be done on those), so I can easily question:

1. The competency (are they measuring correctly?) or truth-fullness of the institutions (are they cooking the numbers?)

2. The relevance of the number (why not just look at people older than X who don't have a job?)

3. The relative importance - maybe unemployment has gone down, but underemployment has gone up, which some would claim is more relevant

These are not even hypothetical concerns, these are all real complaints reasonable people have about the unemployment numbers in various places.

There’s really no reason this would work better than good journalism. The problems we have are that nobody is incentivised to do this (liars make good TV) and even when they are, disenfranchised or just angry people will believe the system is rigged (who settles the bets etc).

Trump Twitter account had everything that went through his head over the last 10 years. It was all public record. He would constantly share lies that would be the entire opposite of what he would have said few years ago. “There’s always a tweet”. Opponents loved to point his contradiction

But in the end? None of that mattered. His partisans didn’t care. What mattered is what he tweeted today. That he said the entire opposite the year before didn’t matter.

you are completely correct, but it's even worse than that:

politicians and partisans across the board have realized that consistency and facts do not matter. hypocrisy has no effect on any politician and their followers. they flip flop position with no hesitation at this point.

Now expand that beyond Trump. You’ll see that it’s commonplace for politicians to say “we need to do X” on Monday then “doing X is terrible” on Tuesday.

Ignoring what politics say is an American (actually global) pastime.

Damn. Your comment really hits the nail on the head, and shows that no matter how much evidence there is of how bad someone is, if the audience’s mind is closed, then the proof is going to fall on deaf ears

Precisely. The whole myth that we are responding to what's factual and what's not is exactly that: a myth, it has no basis in truth. People will believe that which they feel helps their case the most, the facts be damned.

Please HN, feature request ... allow us to upvote comments by a thousand or so. This comment is one in particular which cuts to the heart of the matter.

Best comment from a historian about a certain political party in 1930's Germany.

None of their arguments were arguments in the context of debate. They were post justifications of what they wanted to do. And telegraphing what they wanted done.

Reply on news.ycombinator.com