Biog, on 01 February 2011 - 03:21 PM, said:
People will be able to screen out the good from the bad, by critical thinking and reflection.
When you try something that doesn't work, it is likely to be bad. If you disagree with an author you relied on his work or results but got nothing, you can point it out and warn to rely on it. People will, so, be scared of incredibility if they publish something unreproducible. .
I hate the elitism! I hate Nature and Science magazines and all other elite names or brands, which harvest a lot of money for something that should be done for free.
Uhhh no, try working in a field where there are a lot of papers each week on the topic, if you have the time to read 15+ new papers per week (and catch up on historical papers if you are new in the field) and be able to do bench work full time then you are likely to be a very successful scientist. In my field there are approximately 30000 already published papers on the one protein, not to mention all the other associated papers which are still relevant to the topic, and then there's methodological papers and the new ones at a rate of about 7-8 a week. Without peer-review, the rate of publication would go up (because they could be published straight away), the quality of the publications would go down because people could easily put in papers that weren't complete stories or had major flaws, just to get their publication list up.
Sure, in an ideal world it would be possible to screen out the good from the bad for yourself, but what about the papers where the ideas seem plausible or data has been falsified, but you would have to repeat the experiments to know for yourself... how many experiments can you do and how much money do you have in a week just to verify someone else's data which may or may not be accurate? Have a look at any non-peer reviewed system (e.g. the internet, newspapers) and see how much good stuff is out there, which isn't biased in some manner. Peer review works to screen out these problems to a large extent, though falsified data is difficult.
Nature and Science... well, if you don't like what they do, don't publish with them (though it might be difficult to avoid Nature Publishing Group), but I seriously doubt you will be able to convince many others to not try to publish in those journals, they have the rankings they do for a reason.