Shirley,
Again, I am not against Plos ONE in any way, I think it is a very good idea, but I am (still) not convinced that this will really help measuring the interest of a paper.
People make a point of collaborative thinking as a magic tool. Well, collaboration is certainly good at collecting information (which is the reason why I am blogging and reading blogs…) , but I am not sure that this is so good at changing people’s opinions in a good way. Being a theoretician working in an interdisciplinary field, my day-to-day experience with (established) scientists is that most of them are much more conservative than you would think. One of the reason is that people do not like their old results/ideas to be questioned (and real scientific controversy has indeed almost completely disappeared). Collaboration of conservative people will still make a conservative collaboration, they will just find more papers they like all together 
. Of course there are always more open people, but I doubt they are the majority.
[Maybe that is my French bias; we have this - pejorative- expression in French of "mouton de Panurge", which is the idea that when they start being together, instead of having deeper thinking, people tend to align each other's opinions and completely lose any "out of the box" thinking, kind of "herding effect" . You can see this effect very often in real life, even with very clever people, think for instance of what happens on stock markets. Are scientists more clever, can they all be moderate and keep their originality when thinking collectively ? You can find quite dirty counter-examples on the web -no link - , so I am not so sure... Oh God, I am so pessimistic ! ]