Science Exchange (YC S11) is solving this problem. The basic idea is that instead of the "authorship bartering" model that's pervasive in research today we move towards a market-based collaboration model. Basically we encourage researchers to collaborate because it's financially lucrative to do so. By making researchers wealthy they're able to release themselves from being dependent on prestige publications for generating funding. As a result of this we can start to deprecate the somewhat toxic "public or perish" incumbent system. If your entire livelihood relies on being published in a high impact journal you're incentivized to do anything to make that happen. That is not a system that promotes accuracy of information.