Dan: Good point. I'll try to wriggle out of that one by saying that empiricism is more of a technique than a belief.

Leo: Show 'em the instruments! We weren't expecting the Spanish Inquisition. Then who does?

Sorry folks,I have to bow out of this fascinating discourse as I'm away from the internet for 2 weeks.

I thought some of you might be interested to read this story from The Times Higher Ed http://www.timeshighereducation.co.uk/story.asp?sectioncode=26&storycode=414822&c=1.

Nature is starting a new online journal that is open access with a streamlined review process where readers can judge the importance of papers on the basis of comments, downloads, blog entries, forwards, likes etc. and where papers with negative results will also be published.

Sounds like even those in the publishing establishment understand the force of a lot of what Peter is saying. Remains to be seen how far this sort of publishing venture goes to fixing the issues.

If one assumes, as I do, that the group owning Nature is doing this to make a profit and that smaller peer reviewed journals will be eliminated but supposedly higher quality journals such as Nature will remain untouched then the proposal is reasonable for Nature. It will reduce competition. Sounds good business for Nature but I am unconvinced that it will do much good as I would think that it will put more power into the hands of the few - those who are currently in fashion on the issues. It is crucial to note that Professor Ridd was not talking about AGW but generally and clearly mentioned 2 other important issues for today - Murray and GB Reef.
I hope that I am wrong and that it does some good, but when the powerful set out to destroy competition I lack the blind faith to think that much good will come from that self serving action.
Any more optimistic views? Please!

The "scientific system" is not broken.

It may be subject to corruption here and there, if what the article's author claims is true:

"The journal editor may not be neutral on the issue..."

"Reviewers are also not necessarily neutral and may not be truly critical..."

"Reviewers can block the publication of work antagonistic to their own views..."

"...another scientist who thinks there is a problem with the paper may decide not to try to publish an argumentative comment...They don’t want to cause a fuss and antagonise a potentially powerful group of scientists who could affect their funding and future job prospects."

"The scientist may personally know the original author and would feel that it could destroy a friendship..."

"The scientist may also decide not to try to publish a critical comment because past experience has shown that getting comments published is very difficult..."

"If a scientist is... opposing the conventional wisdom, they will likely find that their reviewers will rate their application poorly and not be funded."

...and so on.

That is not evidence that the system is broken, only that the system may be abused by unscrupulous, timid, money-hungry or incompetent people.

It's like criticizing the laws of cricket for Australia's performance with the bat and ball in the recent Ashes series.

Damn good point, Pericles. I wish I'd made it.

Pericles.
An interesting approach that, in my opinion, should be pushed further. I have said earlier that the word 'broken' was unfortunate in that did not really represent the article as a whole. Be that as it may, we can reasonably assume that even if the system is not 'broken', it is open to and probably is being to some extent corrupted by the methods you quote. It is perhaps not completely irrelevant that Ridd emphasises the fact that the people are human, with of course all the frailty that comes with that condition.
So, what to do?
Possibility one is that we do nothing and 'put up with it'. There are dangers with that; the problems might become worse because there is no penalty and everything is hidden. Confidence in Science itself would be eroded with dreadful long term consequences.
Possibility two is that we try to come up with an approach to try to deal with the problem (assuming it exists as I certainly do).
It it possible to think of Ridd's article as a suggested approach? Whether it is a good approach is open to discussion!
Pericles, we are perhaps in the position where we either suggest a method that might alleviate the problem or give up altogether.
One thing is certain I think. Adopting an attitude that everything is fine and dandy is no good.
I would be most interested to hear any suggestions that you may have that might have to improve the running of Science and as a result raise confidence in it.
Oh, by the way, your cricket analogy has a problem in that we know that performances were poor, the evidence is public. Not true for Science unfortunately.