[jonathankoren]

Different, but not better.

The incentives to game the algo remain. People adapt to the environment.

[kilburn]

That's why mechanism design [1] exists as a field of study. The whole idea of that field is to provide the proper incentives to steer the participants towards your objective. Yes, considering they will try to "game" the system however they can.

I'm pretty sure google could do strictly better (i.e.: better in all reasonable accounts) than they do now if they focused on the users' experience instead of revenue for a couple terms.

[1] https://en.wikipedia.org/wiki/Mechanism_design

[amelius]

> The incentives to game the algo remain. People adapt to the environment.

Perhaps it could work if the algorithm changed its algorithm all the time.

[jonathankoren]

You don’t think Google does this already?

[wyclif]

Paul Graham says Google doesn't want to follow anybody down that road (human intervention in search). But ISTM the problem is that even though they don't, they can just throw a giant pile of money at it if they needed to crush a competitor. Also, VC will refuse to invest in anybody doing it because Google.

[jonathankoren]

PaulG doesn’t know what he’s talking about.

[vintermann]

I disagree, he's a pretty good guide to how VCs think, though not always in the ways he wants.