Coturnix alerts me to the fact that I'm late for a party yet again (as usual). Apparently, Nature is feeling the PLoS competition breathing down its neck. This can be seen in a Nature news article entitled "PLoS stays afloat with bulk publishing" by Declan Butler. In it, Butler describes with little attempt to conceal his arrogance and disdain that PLoS is very successful. According to the article, the means by which PLoS is generating income is mainly by philanthropic subsidies and authors' fees from PLoS One. Butler specifically refers to the ~2500 papers published in PLoS One when he says that Butler is dissing a probably century-old successful marketing strategy: first you convince some people with money that your product (PLoS) is the best since sliced bread (Nature). This step usually costs some money, but if you do it right you get a brand name with a reputation for quality. Once that has worked, you take the same product and sell it more cheaply to the masses and that is when the real money comes in. I'm not an economist and I have no idea if this really was the PLoS strategy, but it might have been. PLoS One could only become a success after the brand PLoS had a solid footing in the science community. And now the entire PLoS venture is working fine because of it and will probably be working fine for some time. So of course, for-profit publishers are getting nervous and so are its employees. Butler's arrogant and demeaning wholesale dissing of basically all the authors and 500 volunteer academic editors at PLoS One has, no surprise there, sent the scientific blogosphere up in flames and the comments on the article reflect this devastating criticism as well.
I only have a few sentences to add to the sizzeling hot discussion all over the place. The article in particular and the discussion in general again raises the old question:

Why, with today's technology, do we still have about 20,000 different 19th century journals around?

Butler's "non-light" peer-review is just a popularity contest anyway, Nature Neuroscience editor Noah Gray (among other colleagues there) admits freely that the toughest obstacle is getting past the editors: It needs to be pointed out here that publishing in these "non-light" journals decides over grants, tenure, promotions and thus peoples' careers and livelyhoods. So one could paraphrase the current system of publishing in science in the following way: If the scientific community were a large corporation, it would be out-sourcing it's hiring and firing to a group of ex-employees who either left the corporation because they didn't like it or were fired themselves. Now how many managers would implement such a system in their company?
Instead, we should have one single, decentralized, publicly accessible database where the current assessment by editors (i.e., the "non-light" component of peer review in e.g. Nature) comes after publication as one of many measures of post-publication review and assessment. The first review should be done by scientists on the science - whatever happens to the paper afterwards is open to debate. I, for one, value the input of professional editors and their expert judgement of scientific newsworthyiness and would not want to miss it.

Conflict of interest statement: I have published in Science and PLoS One; I volunteer as academic editor for PLoS One.