> What's the next breakthrough idea that extracts the good signal from all the noise of a trillion web pages? To me this is going to be manual ranking, but the

> What's the next breakthrough idea that extracts the good signal from all the noise of a trillion web pages? To me this is going to be manual ranking, but the biggest issue with manual ranking is not that you have to do it manually, but how do you know if you can trust users who rank pages. The Wikipedia system is a good start but not enough, and the second issue is how do you prevent its business model to kill its quality