The next two paragraphs were in a private email list where there was then a request that I make the comments public. The situation was my response to another scientist, the topic at hand being the scope of the conspiracy that would be involved in pulling of a hoax that CO2 is a greenhouse gas, etc.. I also talked here a while back about the unity of science:
I think a crucial part of that error is a failure to understand how science works. While you and I (and others) look at it and see masses of scientists from different areas and reach a conclusion, others don't. The extra piece of knowledge we have is that science has to hang together as a coherent picture. If climate people were seriously wrong about the radiative properties of CO2, then CO2 lasers would not work. And so on through a very, very long list. Conversely, if climate types were seriously wrong about CO2's radiative properties, laser specialists would look at the climate work and point to the errors and that'd be the end of the wrong climate CO2 work.
Instead, they take the view that science is story-telling. Laser physicists go along with the climate people because the climate folks are telling a story that the laser folks like, not because there's any particular evidence in favor of it. The "It's a liberal conspiracy", or "They only say this because they want to impose one world government" responses are part of this. The he said -- she said journlistic line is exactly this, as the science is presented as two stories the reader is chosing between. They think the scientists are doing the same thing. (How would they know differently?)
Back to the present:
I'll also mention, in terms of how people could tell what scientists actually do, that John Wilkins is taking ideas on sources for describing how it is that scientists reach conclusions:
How Scientists Think: A Book Proposal
The Scientists Operating Manual
While I'm mentioning John, h/t also to this xkcd cartoon, which captures a certain crowd (an attitude I've occasionally borrowed at least part of) quite well: