Slashdot Mirror
Climategate and the Need For Greater Scientific Openness
The Guardian follows up on the recent news that CRU climate scientists were cleared of scientific misconduct with an article that focuses on how the controversy could have been avoided, and public trust retained, had the scientists made more of an effort to be open about their research. You may recall our discussion of a report from Pennsylvania State University; that was followed by another review with similar conclusions. Quoting:
"The review, led by Sir Muir Russell, does not mention the media. Instead, it examines the reaction of the scientists at the UEA's Climatic Research Unit (CRU) to the pressure exerted by bloggers: 'An important feature of the blogosphere is the extent to which it demands openness and access to data. A failure to recognize this and to act appropriately can lead to immense reputational damage by feeding allegations of cover-up.' The review adds: 'We found a lack of recognition of the extent to which earlier action to release information might have minimized the problems.' Pressure on the scientists, whose once esoteric work creating records of past temperatures had gained global significance, was intense. In 2005, CRU head Phil Jones replied to a request: 'We have 25 or so years invested in the work. Why should I make the data available to you, when your aim is to try and find something wrong with it?' But, the review implies, the more they blocked, the more the Freedom of Information requests flooded in."
I think this demonstrates that the idealized version of the scientific method isn't always followed.
"Why should I make the data available to you, when your aim is to try and find something wrong with it?"
Because that is how science works. Any decent scientist would rather say "here is my data, please help me find something wrong with it."
So all of their private conversations are suddenly public record because they get paid with tax dollars? I'm sorry, but you have no right to take away our privacy just because you are the source of our paychecks.
You will, of course, be demanding accountability in military spending that's equal to what scientist using public funds have now, right? How about starting with the total decommissioning of our nuclear weapons? We spend about 8 billion dollars on each nuclear submarine. Has anyone been asked to present a post-Cold War case for ever having one of those?
What part of "A well regulated militia" do you not understand?
It was only a big deal to the paid US shills, there was no "loss of public trust".
Reasonable people listen to scientific consensus.
Popper's notion of science is, frankly, obsolete. It was already obsolete when I was reading Philosophy of Science in the 1970s. He envisages a world in which falsifying an hypothesis invalidates a theory. But modern science - and this includes quantum mechanics as well as climatology - depends on statistical analysis and probability theory. You could almost say that when Schroedinger and Heisenberg defined the Uncertainty Principle and the probabilitistic Wave Equation, physics changed in a way that obsoleted Popper and the whole Victorian idea of science.
Jones is replying to people who don't want to take large amounts of data and mine them, but to find single errors and then claim that this invalidates the lot. He was actually right to tell them to get stuffed - but, because we live in a world dominated by PR and spin, this was misused against him. You are demonstrating the effect of this - you clearly have never read Popper, but you're trying to use a sound-bite as an argument.
From scarped cliff or quarried stone she cries "A thousand types are gone, I care for nothing, no not one."
No, he just doesn't want a bunch of people funded by exxon-mobil selectively quoting tiny portions of his data to support bullshit positions,
Funnily enough, none of the people who asked for the data were funded by Exxon-Mobil. Its boring how facts get submerged by a straightforward lie.
Uh, except actually they were. It's not even particualry a secret-- take a look at who funds the "Heartland Institute" (Hint: Exxon Mobil). Google the "American Petroleum Institute".
For a while they were even offering a payment of ten thousand dollars to every scientist who published a paper casting doubt on global warming. (They stopped this when it got publicized in the Guardian.)
http://www.geoffreylandis.com
Science isn't a priesthood where you must reach a certain level of trust, experience, or whatever to be allowed in. It is open to all, and all have the potential to contribute. My favorite story along those lines is a 9 year old girl that debunked aura readers. The people said "I can feel your aura!" She said "Ok then you stuck your hands through this partition and I'll put my hand over one of yours, you tell me which." Results were taken and tabulated, readers couldn't do it (did a bit worse than chance actually). It was a complete, valid, experiment, has been referenced later and retested, and an elementary student came up with it.
Now that doesn't mean anyone will have USEFUL commentary, but it doesn't mean that people should be excluded just because they aren't an "expert".
In particular, someone may not be an expert at the given science, but might be an expert at something related that is important. So you have a document on climate and a mathematician wants to examine it. He knows jack and shit about climate, he usually doesn't even know what the weather is. However he knows math inside and out. He goes, examines your research and says "Wait a sec, this is wrong. The math here doesn't work. These numbers do not come out right." He can't analyze the climactic theories, but found out that the conclusion was incorrect because the data had been processed wrong. Or perhaps a philosopher who is very skilled at formal logic and analyzing arguments reads the research and says "Ok hang on, you have a gap in your logic. The conclusion does not follow the premises as stated here." Again he not an expert in the field, but he's an expert in logic.
It is highly important that people of different disciplines be allowed to look at research, in particular when said research is very complex. When you are talking about something that is based off of a lot of math conducted on thousands of points of raw data, that is the sort of thing that is ideal to being in "non-experts" on. Get mathematicians, statisticians, probably some cryptography experts (recognizing patterns in randomness is their thing) to look at the data. They might not be able to understand the climate science, but they can analyze the data and the math and say "This calculation is solid," or "This calculation is incorrect." Looking at the parts of the whose with their given expertise can be as or more valuable than trying to look at the whole thing. The climate scientist might look at the whole thing and say "Ya, all the science fits," but only because they assume all the math is right. If the math is wrong then they might say "Oh, well this no longer shows what it says it does."
The problem is that the tree rings accurately track temperatures until fairly recently, when some of the trees in some of the forests show "divergence," while others continue to correctly track temperature. So the trees do not merely diverge from the temperature record, but from one another.
Obviously, the fact that such divergence can occur limits confidence in the use of tree rings for climate reconstruction. On the other hand, it also shows that such divergence can be detected by comparing results from different forests. The fact that the historical tree ring data does not exhibit this kind of discrepancy between different sets of trees argues against divergence being a problem.
Here is a review that discusses the divergence problem and possible causes.