Can you explain what you mean by potentially corrupt peer review and how the ArXiv and PLoS change this ?

Sorry, I don't always see later posts in a timely manner.

I agree with your second point. If the for-profit journals could accomplish this, then I think that's the way to go. However they don't seem apt to do so, which is why PLoS sprung into place. With ArXiv, the difference is that the author can post the article, and the peer review process begins immediately, much like this forum. I think that fully open 'peer review' in this sense is superior to what we get in the biological or medical sciences.

In principle, having a subset of your expert peers privately judging the validity of a submitted paper pre-press is great, but in practice what often happens is that one or more of those peers can use this as equivalent of scientific 'inside information' to guide their own research, or they in some cases can make esoteric claims against the data. While many, I am sure, are fine playing by the 'honor system', private deliberations over one scientists work, typically by their very competitors, creates unneeded rivalries. If the process were in the open, those same reviewers could air their concerns about the data or conclusions, but their comments would also be subject to 'peer review'.