|
Post by dwerth on Nov 22, 2009 4:21:10 GMT
From one of Phil Jones emails: "So other groups around the world have also entered into agreements. I know this doesn't make it right, but it is the way of the world with both instrumental and paleo data. I frequently try and get data from other people without success, sometimes from people who send me the pdf of their paper then tell me they can't send me the series in their plots." So now you know, data really does get refused due to ownership issues, even sometimes to Phil Jones. There is a problem here. If a GSE (Government Sponsored Entity) or a direct Governmental Entity enters into an NDA with another entity to gain access to its priviledged data, then uses the data for many uses, then it is subject to the FOIA law. If they try and skate around this, then they are honoring a contract, not the law. This is a dicey proposition at best, and should be considered carefully before entering into any binding NDA. However, any NDA that I have seen or signed, has in effect, an escape clause when it comes to legal matters. IE the court requires your testimony or a FOIA request is submitted, then your escape clause from your NDA is then triggered.
|
|
|
Post by Graeme on Nov 22, 2009 4:26:58 GMT
They are failing to release intermediate steps and source code Socold! Seeing as the intermediate steps and source code are not scientific requirements to reproduce work I find it's justifiable for scientists to release it at their discretion. You're correct, normally, but what you've said doesn't appear to be the case in paleoclimatology. To give my understanding of what was going on, the following is a paraphrase of the situation: Claim: "Using X proxies, combined using new statistical methods, we have this wonderful graph showing that the current decade is the hottest in at least 1000 years." Query: "Which proxies did you use, and what are the new statistical methods?" Reply: "Not telling." Obviously, the intermediatry steps and the source code is required to determine which proxies were used and to validate if the statistical methods used were, in fact, valid. Simply asking for other scientists to "work it our for themselves" isn't helpful. Happily, other scientists publish enough information that others can try to validate it for themselves, but how do you check if something is correct if you're not given enough information to work it out? eg. <sarcasm> Given raw data of "1, 1, 2, 2, 5", I have applied appropriate corrections to allow for environmental issues and the result is "2, 2, 2, 2, 2" showing clearly that there is no change in the status quo. There -- you have all the raw data required to duplicate my results. If you believe I'm wrong, please tell me what's wrong with what I've done. </sarcasm> Now, if I had specified details as to what the "appropriate corrections" were and why, then it would be something that others could evaluate, but at the moment there's insufficient information for anyone to check my work. I don't think it's unreasonable given this situation that someone would request my intermediatry steps and the source code of the algorithm I used to produce my results. And, yes, it would be for others to determine if I've made a mistake. Not releasing it because someone might find an error, no matter how small or insignificant, is not good science.
|
|
|
Post by sigurdur on Nov 22, 2009 4:27:00 GMT
From one of Phil Jones emails: "So other groups around the world have also entered into agreements. I know this doesn't make it right, but it is the way of the world with both instrumental and paleo data. I frequently try and get data from other people without success, sometimes from people who send me the pdf of their paper then tell me they can't send me the series in their plots." So now you know, data really does get refused due to ownership issues, even sometimes to Phil Jones. There is a problem here. If a GSE (Government Sponsored Entity) or a direct Governmental Entity enters into an NDA with another entity to gain access to its priviledged data, then uses the data for many uses, then it is subject to the FOIA law. If they try and skate around this, then they are honoring a contract, not the law. This is a dicey proposition at best, and should be considered carefully before entering into any binding NDA. However, any NDA that I have seen or signed, has in effect, an escape clause when it comes to legal matters. IE the court requires your testimony or a FOIA request is submitted, then your escape clause from your NDA is then triggered. And that is precisely why the Dept of Justice in the USA will be involved in this matter. There are a lot of legal ramifications in all of this that pleading ignorance won't cover at all.
|
|
|
Post by trbixler on Nov 22, 2009 4:28:28 GMT
|
|
|
Post by socold on Nov 22, 2009 4:29:59 GMT
From one of Phil Jones emails: "So other groups around the world have also entered into agreements. I know this doesn't make it right, but it is the way of the world with both instrumental and paleo data. I frequently try and get data from other people without success, sometimes from people who send me the pdf of their paper then tell me they can't send me the series in their plots." So now you know, data really does get refused due to ownership issues, even sometimes to Phil Jones. There is a problem here. If a GSE (Government Sponsored Entity) or a direct Governmental Entity enters into an NDA with another entity to gain access to its priviledged data, then uses the data for many uses, then it is subject to the FOIA law. If they try and skate around this, then they are honoring a contract, not the law. This is a dicey proposition at best, and should be considered carefully before entering into any binding NDA. However, any NDA that I have seen or signed, has in effect, an escape clause when it comes to legal matters. IE the court requires your testimony or a FOIA request is submitted, then your escape clause from your NDA is then triggered. Im pretty sure, in the UK at least, that the FOIA have a load of get out clauses specifically to allow confidentiality agreements to be held.
|
|
|
Post by sigurdur on Nov 22, 2009 4:38:50 GMT
There is a problem here. If a GSE (Government Sponsored Entity) or a direct Governmental Entity enters into an NDA with another entity to gain access to its priviledged data, then uses the data for many uses, then it is subject to the FOIA law. If they try and skate around this, then they are honoring a contract, not the law. This is a dicey proposition at best, and should be considered carefully before entering into any binding NDA. However, any NDA that I have seen or signed, has in effect, an escape clause when it comes to legal matters. IE the court requires your testimony or a FOIA request is submitted, then your escape clause from your NDA is then triggered. Im pretty sure, in the UK at least, that the FOIA have a load of get out clauses specifically to allow confidentiality agreements to be held. Not in the USA SoCold......that is why I asked if anyone knew British law. But it seems they took US dollars, and with that comes US law. Like I have said......a huge mess.
|
|
|
Post by trbixler on Nov 22, 2009 4:40:31 GMT
I think i have finally got it, its OK to suppress facts, hide data, dodge FOI, delete files, delete emails, revel in opposing scientists death if there is an out in some contract. Now that is science. Now at least in politics one can expect all of the above, which kind of makes them honest. Did you hear the one about Palin's dog doing tricks........
|
|
|
Post by dwerth on Nov 22, 2009 4:47:28 GMT
Socold wrote: In fact they do. A quick check of the UK FOIA law here: www.opsi.gov.uk/Acts/acts2000/ukpga_20000036_en_4#pt2-l1g41shows that they might be escaping out under clause #41 or #43. However, and this is a major caveat, if they took US money, then they are most definately subject to US FOIA law, which is suprisingly enough, much more rigorous.
|
|
|
Post by socold on Nov 22, 2009 4:59:18 GMT
Seeing as the intermediate steps and source code are not scientific requirements to reproduce work I find it's justifiable for scientists to release it at their discretion. You're correct, normally, but what you've said doesn't appear to be the case in paleoclimatology. To give my understanding of what was going on, the following is a paraphrase of the situation: Claim: "Using X proxies, combined using new statistical methods, we have this wonderful graph showing that the current decade is the hottest in at least 1000 years." Query: "Which proxies did you use, and what are the new statistical methods?" Reply: "Not telling." I understand your point, if you want to know precisely how they got the result you need to know their methods in detail. What I am suggesting though is that it's still possible to publish your own reconstruction using methods and proxies of your own choosing. If you find a different result then this constitutes an argument against their result. The difference between the records would also establishes a rough range and give a rough clue as to the uncertainty. This is why it would be very useful if McIntyre would produce a 1000 year reconstruction using methods and proxies he felt appropriate and publish the results. Even if McIntyre thinks the data is not good enough to make reconstructions he could demonstrate even that by producing two very different records using good methods and proxies. It would not only constitute a good argument but also contribute to knowledge of temperature and uncertainty in it over the past 1000 years. I am just following on in the same roleplay fashion below, but just to let you know because it sounds a little snarky, I am just pretending to be a snarky scientist!: I performed my own analysis on the same raw data using <insert a few paragraphs of method explaination> and found 4, 3, 2, 1. So yes as I found a different result I believe yours is wrong, I don't know why, but then neither does anyone else. Your result isn't going to convince anyone while there is such a disagreement. In fact because I've provided more detail of my methods than you, other scientists will be more convinced by my result. Until you come back with a paper that contains more details (or even some details!) on why your method is better than mine I think other scientists will be reluctant to accept your result. Furthermore I know Dr Example, spurred on by this controversy over these different results, is performing his own analysis as we speak.
|
|
|
Post by dwerth on Nov 22, 2009 5:09:46 GMT
In an ideal world, we would have this reconstruction. However, in order to get this reconstruction, it would basically require state sponsorship, to avoid even the most remote appearance of a conflict of interest should the effort be sponsored by, oh, say the oil and gas companies.
Therefore, until this happens, the most that skeptics of AGW theories can hold is for the raw data and methodologies be provided for public scrutiny.
|
|
|
Post by socold on Nov 22, 2009 5:14:08 GMT
I think i have finally got it, its OK to suppress facts, hide data, dodge FOI, delete files, delete emails, revel in opposing scientists death if there is an out in some contract. Now that is science. These three: "suppress facts", "hide data" and possibly "delete files" I think these relate to the science. These are also the weaker allegations. Some of the emails contradict these allegations. There is good evidence for the other two, but they have nothing to do with science.
|
|
|
Post by Ratty on Nov 22, 2009 5:16:58 GMT
|
|
|
Post by socold on Nov 22, 2009 5:19:12 GMT
In an ideal world, we would have this reconstruction. However, in order to get this reconstruction, it would basically require state sponsorship, to avoid even the most remote appearance of a conflict of interest should the effort be sponsored by, oh, say the oil and gas companies. He wouldn't have to do his own field work and produce his own proxies, he could use existing ones. So he wouldn't require any sponsorship. All he would have to do was to use statistical methods he thought best to build a reconstruction out of them.
|
|
|
Post by Ratty on Nov 22, 2009 5:21:24 GMT
How many Socolds are there?
|
|
|
Post by socold on Nov 22, 2009 5:22:25 GMT
only 1 but we are both tired might go to sleep soon
|
|