No need to squint, one can check the graphs, the Met office in England using HadCRUT3 still found the heat is on:
http://www.metoffice.gov.uk/research/hadleycentre/obsdata/HadCRUT3.html
GIGObuster, thanks for the cite. If you want to call comparing one model to another “testing”, you are quite right. However, it’s not testing, it is intercomparison.
In either case, if you re-read my posts, you will see I was referring to testing as applying V&V and SQA to the models. That is the testing I was discussion, the testing that has not been done. Before we even begin to test the models against the real world, we need to make sure that they are bug-free, that the input data is correct, that the algorithms do not contain errors, and that the approximations actually converge. As far as I can determine, this has not been done, although I’d be happy to be proved wrong.
Of course, the models can get tested against the real world, with or without V&V and SQA. Here’s an example of that kind of testing regarding one of the larger climate phenomena, the Indian Monsoon. Report back and tell us how well the models performed.
I also note that you didn’t comment on the fact that the models disagree with each other by up to 9°F. If that’s “testing”, then they failed the test.
Finally, it seems that you approve the use of V&V for the satellites … why don’t you approve of it for the models?
w.
PS - The CMIP project is not called the CMTP for a reason. It’s not the “Climate Model Testing Project”, it is the “Climate Model Intercomparison Project”, which is a very different animal.
Potatoe, Potato. It is still testing when you check if something does work or not.
Is that ok? Because it is not referenced yet, but I have seen reports of NASA adding V&V to the budget for their climate research.
And you completely missed that that was the test, the ones that failed are either corrected or as I understood from the article abut 3 were not going to be used anymore or were used in the comparison just to check how well the new not adjusted ones worked.
As you miss it, there are more mundane reasons why V&V has not been applied to models, AAICR the V&V people at NASA did not think it was needed, but I have seen a paper were the 2006 and 2007 budget included V&V for the climate models too.
Still, that is IMHO a test, and since physics and chemistry are the main components of climate models, the more I check the more I see that the testing never stops.
http://www.metoffice.gov.uk/corporate/pressoffice/myths/index.html
You didn’t quote what you were referring to, but that document does not appear to contradict what was said about him by the other side. Mann admits that he did not release his source code and doesn’t appear to claim that he released his data promptly.
Sorry, but that does not support your claim that I “have arbitrarily decided that you will only accept evidence of testing on future data”
Do you have anything else to go on? Or will you admit that you have constructed a straw man?
The source cited earlier would seem to dispute that claim.
Cite and quote please? Sorry, but since you invented your earlier claim about the NAS, I don’t have much confidence in your appeals to authority.
Can you please quote the article where it says that release of source code is not standard practice in physical science modeling outside of climate modeling?
So are you claiming that releasing source code means that you forfeit your intellectual property rights completely?
Since nobody has tried to answer this question, I will take a stab. My guess is that Antarctic conditions have not behaved as predicted by these climate models. In other words, the warmers’ list of predictions that have come true is cherry-picked at best.
It’s the equivalent of a psychic giving you a list of 10 of her predictions that have come true. If this is the extent to which AGW has been put to the test, then AGW has not been tested.
In most cases, these scientists are either working for or funded by the US Government. It is not their property. It is ours. We paid them to produce it, and now they won’t release it. James Hansen works for NASA. His top secret, high quality computer code for mangling averages of station data is not his intellectual property. It is yours and mine.
w.
Well, since GIGObuster has decided not to report back on the test that I cited above, I guess I’ll have to.
Modern day computer models don’t outperform the methods developed in 1932 for predicting the Asian Monsoon. To quote from the abstract:
Ooops … I can see why GIGObuster didn’t want to report back to us …
w.
Oh my gosh, I can’t believe it, a bunch of climate modelers willing to actually say in public that climate models are wonderful, there’s a first … well, that does it, it’s the famous "tipping point, that proves your case entirely, I’m convinced now. I’ve seen the light, I recant my heretical beliefs, I now believe that the sun rotates around the earth.
Are you serious? You really think that climate modelers extolling the virtues of their climate models means something? Your faith is touching … next you’ll be telling us that you believe that cigarettes won’t harm you, because the tobacco company executives said so.
The mind boggles …
w.
jshore, why do you think that a climate modeler should disclose the results of all tests, whether good or bad? What’s wrong with just disclosing good results and throwing the bad results down the memory hole?
As you well know, the NSF explicitly disagrees with you on this issue for research that they have funded:
Now, if you think the policy should be different, you have the right to fight for changes of the policy. However, you don’t have the right to make up what the policy is.
As for the policy for people actually working at NASA, this directive is typical gooblygook legalize but it seems to say basically that NASA must protect its intellectual property rights and use careful discretion in deciding when to release software:
Here is more information that I frankly did not have the energy to read.
See my answers in posts #518 and 520.
See, for example, his statements:
You might also want to read the letter from AAAS here. It is also interesting to read what fellow Republican Sherwood Boehlert (Chairman at the time of the House Science Committee) had to say to his colleague Joe Barton when Barton started this witchhunt of Mann et al:
I refuse to play your petty little word games. You are not fooling anyone but yourself with them. The fact is that you have come up with all sorts of excuses not to consider tests of the AGW theory to carry much weight…and chief among them seems to be this issue of future data.
The fact that you can find one source, who isn’t even an atmospheric scientist to dispute a claim does not make it incorrect.
See my post above where I quote the letter from the guy at NSF or see what Rep. Boehlert wrote in what I quoted above.
Implicit in the whole discussion in that article is that codes don’t often get released…that, in fact, employees at national labs or universities don’t necessarily even have the authority to release their codes:
And, as I pointed out, this goes without saying in the industrial context!
At any rate, if you have doubts about this, I encourage you to go to any physics journal and see if most of the papers give the source code for the calculations. And, when you find that they don’t, why don’t you send the authors e-mail requesting it and see what sort of responses you get.
Well, in large part you do. Scientists, particularly modelers, spend a large part of their professional lives writing these codes. If someone can act as a parasite and just take these codes and use them, or modify them a bit and use them, then yes, I think your intellectual property rights have in large part been forfeited.
Ok let’s see . . .
In post 518, you state the following:
In case you haven’t guessed, I’m talking about failure in the context of testing climate models against historical data.
In post 520, you state the following:
What about it do you not understand?
You agree that sometimes climate models are tested against historical data.
You agree that sometimes the climate models don’t do well in those tests.
And you agree that even if the climate model does not do well, the result should be dislosed.
Why is it important to disclose even if the climate model does not test well against historical data?
Why?
None of those statements says WHEN the data was released. Yes or no?
i.e. you know perfectly well that you set up a strawman but you refuse to admit it.
And the fact that you find one source that assumes something is true does not make it correct.
Where in those posts does it discuss the issue of money?
:rolleyes:
Oh really? Then I suppose that JK Rowling has forfeited her copyright interest in Harry Potter by publishing?
Besides, how would Hanson be damaged if somebody had the source code to his climate model, modified it, and used it to make their own model?
By the way, another bedrock principle of science is that you take somebody else’s work and build on it yourself. i.e. act as a “parasite.”
Yes, I’m sure that the NAS, the NSF, and the TVA disagree with me. :rolleyes:
As close to a straw man as you can get, as many have admitted before, extreme weather phenomena like hurricanes are not a good way to check climate models, the resolution is not there. And that is precisely what the Met office meant by: “We now need to provide more regional detail and more complete analysis of extreme events.”
That information was there for you to see and you only decided to ignore it and turn to childish taunting. The evidence even from HadCRUT3 still shows the evidence of the earth warming up. It is indeed more than just faith.
brazil84, I have to conclude after all this time that you are happy drowning yourself with a glass of water.
You are demanding why is important to disclose? That is what scientists do. Why is that the source code is not disclosed many times? That is what lawyers and intellectual property do to scientists.
You need only to bark up the right tree so we can save time.
Why? Why is it important to disclose failures and successes in the context of checking climate models against historical data?
How? What lawyer is preventing Hansen from disclosing his source code? How is the concept of intellectual property preventing him from disclosing his source code?
Well, yes, it’s right in black-and-white that the NSF, the party who was funding Mann’s research, does unambiguously disagree with you. Taking somebody else’s work and building on it does not mean you are entitled to their source code.
Well, I linked to the NASA policy on distribution of software that they produce. I don’t know if NASA would outright stop him from releasing it but, at the very least, it looks like he has to jump through some bureaucratic hoops and deal with some lawyers to get it released.
To be fair, I don’t think that it is only the employers. Scientists too sometimes balk at the idea of releasing their code. It represents a lot of hard work and, once they have produced it, applying it to various problems allows them to publish, which is their lifeblood. Furthermore, code is seldom in very good shape to be distributed to outside parties so releasing it usually entails sprucing it up quite a bit with lots of comments and detailed explanations of how to use it. And, even after that, those who look at it usually come back to you with lots of time-consuming questions and do “stupid things” with it, which may not be that stupid really but were not the way you used it or intended it to be used.
Within my corporation, I think I am a lot less protective over my code than most people. (I am talking here about releasing it to others in the company…not releasing outside the company which, as I noted, I have absolutely no legal authority to do and would likely be fired for doing.) But even I would usually prefer to run it myself rather than give it to someone else to run because in the end it is probably less time-consuming and they are also more likely to do something with it that it wasn’t really meant to do.
Why don’t you tell us since it is you who feels so strongly about this and then we can tell you whether or not we agree with you and why.