3 ms·
The problem is the incentives are all wrong for reporting negative outcomes - basically you get no benefit, but it cost you time. What would work would be to p
by danieltillett 13y ago
The problem is the incentives are all wrong for reporting negative outcomes - basically you get no benefit, but it cost you time.
What would work would be to pay each researcher for every negative result they published and more importantly why the idea failed. Often the failure is caused by something minor that the researcher overlooked when doing the experiment initially - I have saved many a project in the pub, by just chatting informally with people about the problems they are having.