very interesting. Outside of computer science, it will be very difficult to break through existing academic structures, but if you can find a (responsible and efficient) way to do it, the possibilities are very promising. Take medical research for example- all major medical studies are submitted to paper journals (it may be a bit of an antiquated system) and reviewed by experts on the editorial boards. Theoretically, you should be able to get your study reviewed not just by a single expert, but by every expert on that subject regardless of geographical location or journal affiliation. Just some food for thought...

>Theoretically, you should be able to get your study reviewed not just by a single expert, but by every expert on that subject regardless of geographical location or journal affiliation.

This is definitely a direction I would like science.io to go in in the future

I've worked on some medical publications, and you would need a very good rating system. There are all sorts of low-quality manuscripts that uninformed consumers might cling to if its quality (and the quality of reviewers) is not made very clear. This is true about all publications, but especially medical ones where many patients are beginning to look at the sources themselves. Maybe you could even provide two versions of each article; a "professional version" and a summarized "consumer version" for the regular non PhD folk...