undefined | Better HN

0 points__MatrixMan__1mo ago0 comments

Agreed, as long as it's a catastrophe that the bettors can't cause, but for which advance warning can mitigate harms.

For instance, I'm in favor of bets that a certain astroid will strike the earth at a certain time and place. A signal from the prediction markets might cause somebody to evacuate in a scenario where they'd otherwise cry "fake news."

Let's not bet on whether the water will remain drinkable, because the last thing we need is for somebody to have an incentive to poison it.

0 comments

jacobgkau1mo ago

> For instance, I'm in favor of bets that a certain astroid will strike the earth at a certain time and place. A signal from the prediction markets might cause somebody to evacuate in a scenario where they'd otherwise cry "fake news."

I understand the point you're making, but in this case, you're still incentivizing someone somewhere to not attempt to the best of their ability to intervene in that astroid. Bets that truly can't cause any change in behavior that might affect the outcome are a mostly theoretical category, in my opinion.

__MatrixMan__OP1mo ago

If the bet can't cause any change in behavior then the whole thing is useless. The whole point is to do some good with it. The constraint is that the bettor can't alter the outcome.

Another one would be discovering malware in a PR and betting loudly enough that it won't get merged. The bet is how you make your certainty rise above the bot noise and attract extra attention on the maintainers' part.

Granted that's also theoretical, but it's worth theorising about how we'll get things done in a world where the only way to be heard is to put your money where your mouth is.

jacobgkau1mo ago

> Another one would be discovering malware in a PR and betting loudly enough that it won't get merged. The bet is how you make your certainty rise above the bot noise and attract extra attention on the maintainers' part.

This example seems weaker than the asteroid example. Consider that if you're betting "loudly enough" it won't get merged, the less likely option becomes the malware actually being merged. Now you have a repo maintainer who can bet that it'll get merged (probably a more lucrative bet since it's less likely), and merge it to make money from that bet.

j / k navigate · click thread line to collapse

0 comments

jacobgkau1mo ago

__MatrixMan__OP1mo ago

If the bet can't cause any change in behavior then the whole thing is useless. The whole point is to do some good with it. The constraint is that the bettor can't alter the outcome.

Granted that's also theoretical, but it's worth theorising about how we'll get things done in a world where the only way to be heard is to put your money where your mouth is.

jacobgkau1mo ago

j / k navigate · click thread line to collapse