= 27d567bc2d48d9f40e9ccc20ea370b15)
Lumifer comments on Open Thread, May 12 - 18, 2014 - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (201)
So you are suggesting a system that relies on the opinions of people who got selected because they really want to see their names at the top of the page in big font?
If the only way to see their names at the top of the page in big font is to provide correct predictions... why not?
The classical prediction market relies on opinions of people who got selected because they really wanted to make money on the prediction market. What's the big difference?
Okay... I can imagine that if someone's goal is to bring attention to themselves, they might make correct predictions to get to the top, and then intentionally make shocking (incorrect) predictions to bring even more attention to them. Kinda like people with too much karma sometimes start trolling, because, why not.
Money is a MUCH better motivator.
In particular, making predictions is not costless. To consistently produce good forecasts you need to commit resources to the task -- off-the-cuff opinions are probably not going to make it. Why should serious people commit resources, including their valuable time, if the only benefit they get is seeing their name in big letters on top of a long list?
Well, this is something that can be tested experimentally. It could be statistically tested whether the results of the top predictors resemble random noise.
Some people spend incredible amounts of time on internet, reading about stuff that interests them. I can imagine they could make good predictions in their area. (And there should be a "no vote" option for questions outside of their area.)
Wikipedia exists, despite it doesn't pay its contributors, unlike other encyclopedias. And there is some good stuff there. Also bad stuff... but that's what the competition between predictors could fix.
There is probably a limit on how difficult things can be predicted. But it could be higher than we imagine. Especially if the predictions become popular, so for many topics there would be predictors whose hobby it is.
There are some technical details to solve, e.g. whether the predictor's prestige will be global, or topic-dependent. (To prevent people from systematically giving 10 great prediction in topic X, and then 1 bad but very visible in topic Y.) But that's like having multiple StackExchange accounts.