Overall, this is correct. The problem is that there were some journals owned/operated by some very well-known publishers, such as Elsevier. It seems those publishers have mostly taken action to immediately close down their less reputable journals, but it still remains that you have to be particularly careful when dealing with open access journals.

@YJ - I have never suggested that "peer review" is meaningless. But peer review in no way equals "100 % accurate and unbiased" - which is what I have suggested many times and presented evidence to support. Abge is only supporting my past claims in a different manner.

If a bunch of bought-and-paid-for “research” scientists review your paper – and said paper is say a corollary or closely tied to your research (for which you are highly compensated to conduct…) there is a good chance that your review of his work is going to be biased positive because, you know, if his research is fraudulent, then yours might be, too.

So of course there is no guarantee of integrity in peer-review.

This does not mean there is no value to peer review – on the contrary, I believe it does add value – but it is not a guarantee of anything.

Fair enough.

yj - damn you and your agreeableness...

Not too sure about the title of the thread - it seems a little prejudicial and generalist to me. Anyway putting that aside, there is another point of view:

''The real problem for science today is quality control. Peer review has been at the heart of this, but there are too many failures – both in open access and traditional journals – simply to plod ahead with the same system. We need new approaches and numerous individuals and organisations are working on these, such as the open evaluation project.

The creative potential offered by digital communication of scientific results, an area in which open access journals are leading the way, is exactly where we need to focus. And if we do so, we will solve the problem of the broken peer review system that Science and the gonzo scientist have uncovered.''
http://www.theguardian.com/higher-education-network/blog/2013/oct/04/science-hoax-peer-review-open-access

You can not eliminate fraud and bias when humans are involved.

I'm a huge supporter of the proliferation of science literature and I think open access journals have a large part to play in that. People complain about the peer review process and yet no one seems to be able to come up with reasonable alternatives. Open Access Journals, however, seem to be particularly vulnerable to corruption.

Even if one were to accept your premise that open access journals are more prone to fraud the ever more prohibitive monetary cost of closed journal subscription remains. Indeed, the article I referenced made mention, I think, of Harvard having to cut back on their e library resources. I wonder what might be the greater evil - the possible heightened susceptibility of open journals to fraud or the ever more restricted access to scientific research afforded by the status quo. One thing that might well help is that all journals, open or closed, stipulate that all raw data is presented, perhaps as an internet addendum. Another might be that grant giving bodies include an allowance for open access publication in their costing. In any event it seems wrong, on a fundamental level that publically funded research is not made available for that same public to freely read.
I have in mind in particular the reluctance of the Hadley centre to publish their raw data or even make it available when other researchers request it.

"In any event it seems wrong, on a fundamental level that publically funded research is not made available for that same public to freely read."

I could not agree with you more.

However, I think I would argue that inaccurate science is much more damaging than locked down science. The reality is, if you really need to find an article, you can get it. But, the proliferation of bad science that looks legit could be extremely damaging to real publications' credibility as well as science literacy in general.

Open access journals are the scientific equivalent of vanity presses for regular authors. Peer review is somewhat better, but there has been enough fraudulent scientific papers published there too.

Releasing raw data only works when that hasn't been faked too and with computers that's easy to do. All you need to do is run a program that generates data to fit the desired result and add a random element to each point to make it look real.

‘’ However, I think I would argue that inaccurate science is much more damaging than locked down science. The reality is, if you really need to find an article, you can get it. But, the proliferation of bad science that looks legit could be extremely damaging to real publications' credibility as well as science literacy in general.’’

I am not sure that this is an open or subscription journal issue. May I point you in the direction of the ever expanding C.PubMed database? It seems to form a pretty workable template to me. Here is the FAQ page in case you are interested:
http://www.ncbi.nlm.nih.gov/pmc/about/faq/#q9

‘’Open access journals are the scientific equivalent of vanity presses for regular authors. Peer review is somewhat better, but there has been enough fraudulent scientific papers published there too.’’
Peer review is very much part of the best open access journals (e.g., see my previous post on C.PubMed)
‘’Releasing raw data only works when that hasn't been faked too and with computers that's easy to do. All you need to do is run a program that generates data to fit the desired result and add a random element to each point to make it look real.’’
I agree that a case of outright fraud is possible via this data manipulation. However, at least, if the data is published, one has the opportunity to repeat the experiment and confirm or not the datas’ validity. I, however, am much more interested in bias. That is how the authors interpret the data and how statistically valid, or otherwise, their iterpretation is (e.g., historic tree ring data used for climate change proofs). It might also be that honest arithmetical mistakes have been made. Whether it is fraud or bias or plain old adding up mistakes, the wider scientific community therefore gets to perform the function of almost a super peer review corpus if required. Surely that can be no bad thing?