Reading through the retractions and "editorial concerns" from major scientific publishing houses got me thinking about how the internet has compromised the dissemination of research information. It is usually thought of going the other way - the internet aids in the dissemination of information, but for scientific research this is bad news. Here's why:

For centuries scientists have written up detailed notes about their experiments - the variations they attempted, the observation they made, the results they obtained - then they let other scientists read them, to comment on it. 

This then fuels the next set of experiments, spurs others to repeat the previous for verification and around and around until a working knowledge base has formed. This system allows commentary on the commentary, which serves to strengthen the knowledge base by eliminating weak elements.

This system is still present in the modern system of peer review. However, with the internet,  every little result and published paper is disseminated to the public. Yet, these tidbits are not yet "verified" - yes, they went through peer review, but they haven't had those extra layers of commentary before reaching the public. The public IS the extra layer of commentary, and that gives every element, no matter how weak, equal standing. 

The knowledge base is getting lopsided. 

Some may argue that retractions make science look bad, but really it's the process of weeding out the weak arguments and data. In the age of the internet publishers may need to add in an open review of peer review before publishing anything simply to take back the system of dissemination. Just because a finding was published doesn't make it fact - it takes many identical findings and consensus before that occurs. But you'd never know it from reading the headlines.