Climategate • Hacker News Board

Re: Climategate

#21

Post by gjm11 » Thu, Nov 26, 2009, 9:04 PM UTC

1gor wrote: post #7

gjm11 wrote: post #5

Earlier quoted context omitted.

Er, no, he doesn't say that the main climate model is useless, or that "it's rubbish and doesn't work". He doesn't say anything even slightly like that. The particular thing he's saying is "meaningless" is a figure that has basically nothing to do with actual climate simulation. Clearly he's not impressed with the quality of the code. Fair enough. There's a lot of very poor code around doing important things. It look…

No, you are wrong. Read thee whole quote here http://www.cbsnews.com/blogs/2009/11/24/taking_liberties/ent... and observe that the programmer is convinced the whole database of reporting stations is bogus, and the model has been tweaked multiple times: "Each parameter has a tortuous history of manual and semi-automated interventions that I simply cannot just go back to early versions and run the update prog." also "O…

I read not only the whole quote but the whole (very long) file from which the quote was cherrypicked.

Of course the model has been tweaked several times. What the hell else do you think scientists do when they have a simulation of a large complicated system and there's a constant influx of data for seeing how well it works?

The fact that they have to work with messy data is hardly their fault.

Re: Climategate

#22

Post by gjm11 » Thu, Nov 26, 2009, 9:09 PM UTC

mixmax wrote: post #11

gjm11 wrote: post #5

Earlier quoted context omitted.

Er, no, he doesn't say that the main climate model is useless, or that "it's rubbish and doesn't work". He doesn't say anything even slightly like that. The particular thing he's saying is "meaningless" is a figure that has basically nothing to do with actual climate simulation. Clearly he's not impressed with the quality of the code. Fair enough. There's a lot of very poor code around doing important things. It look…

My point was that it seems like the model might be flawed, or at least is poorly written and/or documented. As you say, this Ian guy is clearly not impressed with the code. This qoute describes it rather well: Ulp! I am seriously close to giving up, again. The history of this is so complex that I can't get far enough into it before by head hurts and I have to stop. Each parameter has a tortuous history of manual and…

Whatever makes you think there is only one global climate model?

Re: Climategate

#23

Post by yummyfajitas » Thu, Nov 26, 2009, 9:45 PM UTC

gjm11 wrote: post #21

1gor wrote: post #7

Earlier quoted context omitted.

No, you are wrong. Read thee whole quote here http://www.cbsnews.com/blogs/2009/11/24/taking_liberties/ent... and observe that the programmer is convinced the whole database of reporting stations is bogus, and the model has been tweaked multiple times: "Each parameter has a tortuous history of manual and semi-automated interventions that I simply cannot just go back to early versions and run the update prog." also "O…

I read not only the whole quote but the whole (very long) file from which the quote was cherrypicked. Of course the model has been tweaked several times. What the hell else do you think scientists do when they have a simulation of a large complicated system and there's a constant influx of data for seeing how well it works? The fact that they have to work with messy data is hardly their fault.

The fact that a sum of squares yields negative numbers is their fault. The fact that they can't reproduce their own numbers is also their fault.

Re: Climategate

#24

Post by jswinghammer » Fri, Nov 27, 2009, 2:47 AM UTC

catweasel wrote: post #17

barry-cotter wrote: post #13

Earlier quoted context omitted.

Extremely debatable. Better regulation would have prevented the housing crash but the regulation Americans got (with bipartsian support!) had the effect of making borrowing to buy a house ever easier, thus inflating prices, leading to a crash when it became obvious there were no greater fools left.

Which specific pieces of regulation made borrowing easier? An interesting wiki on the subject is the GFC timeline- http://en.wikipedia.org/wiki/Subprime_crisis_impact_timeline There's mention there of a political goal ... "of increasing minority home owners by at least 5.5 million by 2010 through billions of dollars in tax credits " ... which is not the same as regulation. The timeline does attribute removal of regul…

The Community Reinvestment Act encouraged banks to make risky loans and if banks didn't make such loans the government would start snooping around in their business. The understanding has always been that the government would bail out massive losses in these loans. Some of the deregulation made things easier for this to happen but the recession would have been regardless of these factors. Job growth over the last ten years has been almost entirely focused in housing and related industries.

Re: Climategate

#25

Post by jswinghammer » Fri, Nov 27, 2009, 2:49 AM UTC

catweasel wrote: post #6

jswinghammer wrote: post #4

Earlier quoted context omitted.

It's interesting that this information isn't offered on articles describing emissions caps that are being agreed to in advance of these upcoming talks on climate change. I'd imagine that from now on other climate research departments will keep a tighter lid on their internal communications at any rate. I never really thought I'd miss Bush but at least he was smart enough to not take the global warming stuff seriously…

We're not in a depression, we're in a recession.. and it was the lack of regulation that caused it not abundance.

If you count underemployed, discouraged workers and the U-2 figure then we're at about 18% unemployment right now. If we're still in this situation after ten years despite all the artificial stimulus I think it's just a joke to call it anything less than a depression. I'd argue we've been in a depression since the tech bubble burst and that we've just been papering over the mess for the last ten years. Considering that job growth has been basically zero for that period I'd say we've been kidding ourselves with statistics for awhile now.

Re: Climategate

#26

Post by catweasel » Fri, Nov 27, 2009, 3:12 AM UTC

jswinghammer wrote: post #24

catweasel wrote: post #17

Earlier quoted context omitted.

Which specific pieces of regulation made borrowing easier? An interesting wiki on the subject is the GFC timeline- http://en.wikipedia.org/wiki/Subprime_crisis_impact_timeline There's mention there of a political goal ... "of increasing minority home owners by at least 5.5 million by 2010 through billions of dollars in tax credits " ... which is not the same as regulation. The timeline does attribute removal of regul…

The Community Reinvestment Act encouraged banks to make risky loans and if banks didn't make such loans the government would start snooping around in their business. The understanding has always been that the government would bail out massive losses in these loans. Some of the deregulation made things easier for this to happen but the recession would have been regardless of these factors. Job growth over the last ten…

An act which may encourage banks to lend is not regulation. The belief that gov't would bail out the big players does not indicate over regulation, or even appropriate regulation. Quite simply it was the regulation exemptions that were provided which made this thing far worse than it could have been.. and far more widespread. The idea that large financial institutions would conduct themselves responsibly without regulation, and not give way to greed, is complete rubbish and has been proven so, even the most conservative on capitol hill admit that these days.

Re: Climategate

#27

Post by miked » Fri, Nov 27, 2009, 3:13 AM UTC

gjm11 wrote: post #5

mixmax wrote: post #2

The real scandal is more subtle, and much more serious. There is evidence from some of the leaked e-mails and source code that the main climate model is, in the words of an unnamed developer called Ian, useless: I am seriously worried that our flagship gridded data product is produced by Delaunay triangulation - apparently linear as well. As far as I can see, this renders the station counts totally meaningless. It al…

Er, no, he doesn't say that the main climate model is useless, or that "it's rubbish and doesn't work". He doesn't say anything even slightly like that. The particular thing he's saying is "meaningless" is a figure that has basically nothing to do with actual climate simulation. Clearly he's not impressed with the quality of the code. Fair enough. There's a lot of very poor code around doing important things. It look…

Er, no, he doesn't say that the main climate model is useless, or that "it's rubbish and doesn't work". He doesn't say anything even slightly like that.

It doesn't make any difference what judgements he pronounces. What matters most is his factual observations. At one point he notes that a sum of squares routine sometimes produces a negative result(!). He also notes that they are not able to reproduce their earlier results using the model. Reproducibility is one of the foundations of science. Not only can others not reproduce their results, they can't even reproduce their own results.

This raises a further question that I haven't seen mentioned: was this code and data under version control? And was that lack of version control considered a bug or a feature? We are talking about billions of dollars being spent based on models that apparently had no source control . If they did, they should be able to reproduce.

Re: Climategate

#28

Post by catweasel » Fri, Nov 27, 2009, 5:32 AM UTC

catweasel wrote: post #26

jswinghammer wrote: post #24

Earlier quoted context omitted.

The Community Reinvestment Act encouraged banks to make risky loans and if banks didn't make such loans the government would start snooping around in their business. The understanding has always been that the government would bail out massive losses in these loans. Some of the deregulation made things easier for this to happen but the recession would have been regardless of these factors. Job growth over the last ten…

An act which may encourage banks to lend is not regulation . The belief that gov't would bail out the big players does not indicate over regulation, or even appropriate regulation. Quite simply it was the regulation exemptions that were provided which made this thing far worse than it could have been.. and far more widespread. The idea that large financial institutions would conduct themselves responsibly without reg…

This juvenile practise of downvoting simply because one disagrees does not change the fact that legislation != regulation, in fact one can legislate to remove regulation. And btw, the Community Reinvestment Act was introduced in 1977, I doubt it played a major influence in the GFC of 2008. In 2000 regulation was removed for credit default swaps, Lehman, AIG, Bear-Sterns,FM & FM and others all went under due to overload of toxic credit default swaps. Proper regulation could have avoided that... now we have the undesired result of Government owning some of these failed creatures.. which is even worse than government regulations.