Ask HN: Start formal group to peer review climate code? | Hacker News Reader

Ask HN: Start formal group to peer review climate code?

31 pointshop39 comments

Wanted to see if some people had interest in forming a formal group to vet algorithms used to calculate climate data. In the wake of the CRU emails that included questionable programs written to interpolate temperature data, I think there will be a need for an independent review by coding experts.

Maybe a wiki style project with everything laid out in the open.

For the record, I'm impartial and agnostic about climate change because I haven't seen the raw data or the programs used to make future predictions. I think this whole "consensus of scientists so its right" is BS and goes against the scientific method. Plus, climatologists may be incentivised to spread FUD b/c it brings fame and more grant money - plus their expertise is not statistics and writing code.

A software engineer reviews CRU source code - http://www.youtube.com/watch?v=q_buKNBrpcM

I'm sure some funding would not be difficult to find, given the magnitude of all this.

ClimateCodeReview.org? ClimateHackers.org? ClimateCrunchers.org?...

View on HN

One thing I don't get is how people always seem to follow the pattern "I'm unbiased. #{strong_opinion_here}."

Unbiased... really, hop?

For the record: I think that open source + code cleanup + code review is a great idea, and not just for climate science. But it's important to keep in mind that ugly code isn't the same thing as wrong code, and that mistakes in the code aren't automatically become mistakes in the science.

(For example, a poorly written data parser with lots of bugs will not affect the science one bit, if none of the data triggers those bugs.)

So what's your purpose here:

- To contribute to a community, finding and fixing bugs, making things more readable and robust, etc?

- To do a little audit to "prove" what you already believe? (I suspect a hint of witch-hunt in the air...)

Edit to fix formatting and add a side comment: This "climatologists may be incentivised to spread FUD b/c it brings fame and more grant money" bit doesn't pass the smell test for me. If this is the true motivation of climatologists, then why aren't more of them going for the fame and think-tank money available to them by expressing dissent? I can't believe that all these scientists are dishonest and greedy enough to lie about the science, but not suitably dishonest/greedy to sell out their peers.

It's possible that the post has been edited since you replied, but as it reads now he is "impartial and agnostic about climate change". Clearly he is biased against the current state of "climate science" as currently practiced, but I'd take him at his word that he's interested in seeing what the data show if one removes the hype and spin.

I am biased - not about climatology, but having been in grad school and seen it first hand, I am biased about scientists motives, especially when you are competing for millions of dollars of grant money. I also know its easy to fit big data sets to a desired conclusion.

I care about being confident that the data supports the conclusion. I want all the data in the open and open to scrutiny. Whether the Earth is cooling, heating, staying the same - I just want it proven with open source code, not on a consensus of people I don't know using data hidden from the public, with code that I haven't read.

Edit: I am sure the vast majority of scientists are true to the data. And this project will only help reinforce their work.

Even if you could get ALL the data in an open and transparent format and were able to scientifically prove that "Global warming is (100% real/fake)" it wouldn't matter because the political/corporate/celebrity forces on the other side would claim that you are "using bad science, corrupt, in the pocket of (liberal elites/big corporations)". Even if you proved that there isn't enough data to prove anything, both sides would say you are wrong.

It's still worthwhile.

A grab-bag of replies before I get back to work. (Yes, on a Sunday; sigh.)

- The notion that science can be "proved" is a mistaken one.

- I don't think there's much question by anyone that there has been measurable warming this century; so that part of your question is answered. ;)

- How much of the data is really "hidden"? There certainly hasn't been a effort to widely distribute it from a central source (until lately: http://www.realclimate.org/index.php/data-sources/), but that doesn't mean it was under lock and key either.

Personally, I am biased in the other direction: I feel that it is likely that the extreme pressure and politics involved are forcing climate scientists to be over-conservative in their conclusions.

I am biased about scientists motives, especially when you are competing for millions of dollars of grant money.

You raise legitimate points and I think the idea of open-source climate modeling (perhaps using BOINC or something for distributed processing) is an excellent one.

Unfortunately, I think the value of your proposal is obscured by comments such as the one above. Surely, scientists do compete for millions of dollars in grant money, but I'm also a bit biased about vested interests that are competing for billions (or perhaps even trillions) of dollars in revenue from the sale or processing of fossil fuels, with any inconvenient externalities dumped on the taxpayer. To be sure, such fuels have been essential progress and are an important factor of future economic growth, but suppliers are not necessarily altruistic either.

He said, "I'm unbiased on issue X, but I strongly believe that Y group of people can not be trusted about X."

The only reason you read this as "I have a strong opinion about X" is because the primary argument in favor of X consist of "trust Y on the issue of X."

(By "primary argument", I mean within the mainstream media. Within the scientific community, there are both legitimate scientific arguments and illegitimate attempts to subvert science via politics.)

>"This "climatologists may be incentivised to spread FUD b/c it brings fame and more grant money" bit doesn't pass the smell test for me. If this is the true motivation of climatologists, then why aren't more of them going for the fame and think-tank money available to them by expressing dissent?"

Many people go into environmental science because they are environmentalists. You get the same results as when you send a group of Catholic priests to investigate the virgin birth.

Or, picking on my own biases, it is like sending a bunch of libertarian economists to estimate the utility of a government program.

Also, the billions that governments are spending on climate change policy and research easily dwarfs money on the other side. That's a bit of FUD spread by the alarmists that doesn't pass the smell test.

Trust, but verify.

> Also, the billions that governments are spending on climate change policy and research easily dwarfs money on the other side. That's a bit of FUD spread by the alarmists that doesn't pass the smell test.

Well, this looks to resemble a checkable fact, rather than the supposition and ad homeneim the rest of your post was filled with.

Data point: In 2006, Exxon alone spent 733 million in R&D (from their annual report). That's about 5x (edit: I had written 6x) the federal National Center for Atmospheric Research's budget. (153 million) (source: http://www.nsf.gov/statistics/nsf08318/content.cfm?pub_id=39...)

Data point: The American Enterprise Institute's 2006 income: 28.4 million. (From sourcewatch.org) Greenpeace USA? 11 million. (from the 2006-7 annual report)

Data point: Environmental & health lobby spending is dwarfed by that from energy groups: http://www.publicintegrity.org/investigations/climate_change...

Data point: Anti-climate research for cash: http://www.guardian.co.uk/environment/2007/feb/02/frontpagen...

I think that's enough for now. To claim that the "other side" is dwarfed is laughable, at least as long as you are speaking dollars and not facts or mindshare. At the _very_ least, it's an even fight (looking at research dollars); more likely the resources on the anti-climate-change side far outweigh those of scientists and climate change activists (including policy & advocacy dollars).

Edit: tsk, tsk. Reply, don't downmod!

Exxon's R&D budget isn't going into global warming research - it mostly goes into research into how to find and tap oil (geology, physics, mathematical modeling, etc). A substantial portion (20-30% - from what I can gather) goes into green research (biofuel, solar, fuel cells, carbon capture, etc).

You can't compare the 'income' of two arbitrary non-profits (greenpeace and AEI) and draw any conclusion from it either. Seriously, it's idiotic and misleading to try to do so.

And NCAR doesn't just do global warming research, either.

There really is no such thing that I can see as "anti-climate-change" research so it would be rather difficult to find a direct comparison.

> You can't compare the 'income' of two arbitrary non-profits (greenpeace and AEI) and draw any conclusion from it either.

Why not? My purpose here is only to give a counterexample to the 'policy' portion of the claim that the "anti-climate-change" side is "dwarfed" by the pro- side. It's fair to say that both Greenpeace and AEI do are quite active on the issue (I picked the first two that came to mind); equally it's fair to say that Greenpeace certainly does not 'dwarf' AEI; and further it seems reasonable that Greenpeace isn't massively outspending AEI even on the single issue of climate change. It seems to me almost certain that the opposite is true.

Exxon spends a ton on R&D, but very, very little of it is for climate science research. As far as I can tell, they're total amount of spending on climate science research the past few years was about $75,000 ( http://members.greenpeace.org/blog/exxonsecrets/2009/05/26/e... ).

The American Enterprise is not a climate research institute, it's a general purpose, conservative policy shop. It does not fund original climate research, it just funds policy experts who write issue papers. It's comparable institute on the left is the Brookings Institute ( which incidentally has a budget almost three times the size: http://en.wikipedia.org/wiki/Brookings_Institution ).

As far I can tell sum of the amount of money on the skeptic side, for actual climate researchers, consists of Steve McIntyre's retirement fund and a the salaries of a few tenured professors. On the other side, there are about 2,500 scientists who get their money through grant funding, and the people in charge of dispensing those grants are either governments or other non-skeptical, climate scientists. The overwhelming amount of the money goes to organizations and scientists who believe in AGW.

I took a broader view of "the other side" from the grandparent comment. If you're going to assert that within climate research itself there are very few dissenters; well, yes. That's what having a broad scientific consensus means, after all!

On the advocacy/policy front, 'the other side' is more well-defined and it seems clear to me that claims for it being dwarfed are incorrect.

(AEI vs Greenpeace was just an out-of-the-hat comparison of two organizations that advocate loudly on opposite sides of the issue; you can look at the lobbyists-by-industry chart for a less randomly-chosen view of the same thing.)

The problem is that the grant funded climate scientists all get more funding if they raise more alarm about global warming. Even more concerning, the distribution of grants, the peer review process, and the IPCC process are controlled by people like Mann and Phil Jones who study the issue like lawyers prosecuting a case, not like scientists proving a case. As a result, new scientists who are skeptical have little chance of getting grants, getting published, or getting tenure. Climate science has the same selection of the process as a priesthood, and that kind of selection process will produce the same errors as a religion. But right now, there is no alternative source providing funding. So all we hear about is a "consensus" when that consensus is really just a self-replicating group of believers.

> The problem is that the grant funded climate scientists all get more funding if they raise more alarm about global warming.

This is often asserted, but is it true? Most alarmism that I have seen is coming from the media or activist organizations; the few times I've seen scientist alarm has been when recent measurements have been worse in some way than their models predict.

Further, let's assume the funding relationship is true. Are all these scientists that dishonest? Wouldn't honest ones be reigning in their colleagues? Why haven't we seen this?

> Even more concerning, the distribution of grants, the peer review process, and the IPCC process are controlled by people like Mann and Phil Jones who study the issue like lawyers prosecuting a case, not like scientists proving a case.

They're expected to present and defend their claims in the court of public opinion and in the realm of politics; why are you shocked and appalled that they are doing so?

N.b. they're rather incompetent at public relations, unfortunately...

> Climate science has the same selection of the process as a priesthood, and that kind of selection process will produce the same errors as a religion.

Sure. So does astrophysics, evolutionary biology, ... How is the situation different from any other deep speciality?

> So all we hear about is a "consensus" when that consensus is really just a self-replicating group of believers.

Or, a consensus opinion under attack by politically motivated non-scientists. What test could be applied, or evidence provided, to tell the difference?

How is the situation different from any other deep speciality?

I've never heard of influential people in quantum mechanics trying to get a journal shut down because it published something critical of many worlds.

Or, a consensus opinion under attack by politically motivated non-scientists. What test could be applied, or evidence provided, to tell the difference?

We could, I dunno, hack their email to try and see if they try to suppress dissenting opinions?

We could also try to examine the science to see if it truly is as solid as they argue. Well, actually we can't, since they deleted the data (oops, I mean lost it accidentally, they never acted on the emails in which they said they would delete it)...

> I've never heard of influential people in quantum mechanics trying to get a journal shut down because it published something critical of many worlds.

You're making a poor choice by using 'many worlds', which is as much/more about philosophy as it is about physics. But if you're looking for blacklisting, banning, etc, here's an example for you, specifically from influential people in quantum mechanics: Arxiv blacklists topics/people all the time. http://www.scientificblogging.com/quantum_diaries_survivor/d... http://www.archivefreedom.org/ http://archivefreedom.org/freedom/Cyberia.html

> We could, I dunno, hack their email to try and see if they try to suppress dissenting opinions?

If what's been brought to light consists of anything more than angry muttering I'd like to see it. Personally I'd say the anger is more than justified, although it was stupid to bring it out of the bar and into public record.

> Well, actually we can't, since they deleted the data (oops, I mean lost it accidentally, they never acted on the emails in which they said they would delete it)...

It's my understanding that the 'lost' data is (a) available from original sources, since it was a compilation of weather station data; (b) still fully available in 'cleaned-up'/scrubbed form (e.g. deleting data that was duplicated from multiple sources), (c) not the data referred to in the hacked emails (it was about other emails, no?).

If you're determined to find a conspiracy, you will succeed -- just as intelligent design types have invented a conspiracy of evolutionary biologists and new-age moonbats have invented a conspiracy of physicists.

Personally I have a hard enough time believing in conspiracies when the alleged participants are disciplined and collectivist, let alone when they're a bunch of egotistical scientists.

(Aside: are you really suggesting a vigilante hacking war is acceptable? If so, I think the conservative lobbyists on K street would be an excellent place to strike back; maybe we'd even find out which of them hired the hacker. (See, I can do conspiracy theory too!))

I can only comment on Carlos Castro, the link of yours at archivefreedom.org. Near as I can tell, he claims to have come up with a unified field theory. As far as I know, the arxiv filters out proofs of P=NP, the Riemann Hypothesis, unified field theory and similar things. So what?

That's not the same as trying to get JMP shut down or editors ousted because it publishes papers on Bohmian mechanics.

The fact is, Steve McIntyre is not a crackpot. He has made important contributions to the field, most notably making corrections to Michael Mann's "hockey stick". The CRU folks conspired to keep data from him in order to prevent him from making further corrections to Briffa's hockey stick.

With regards to angry muttering, there is far more than that. There is conclusive evidence of a conspiracy:

And don’t leave stuff lying around on ftp sites – you never know who is trawling them. The two MMs have been after the CRU station data for years. If they ever hear there is a Freedom of Information Act now in the UK, I think I’ll delete the file rather than send to anyone.

Can you delete any emails you may have had with Keith re AR4?

With regards your point c), the people who conspired to hide data now claim that other data was lost. And we believe them because they are so trustworthy?

With regards to a hacking war, I wasn't advocating it. I'm simply pointing out that the recent incident does give us evidence that the consensus (which is not a substitute for solid science, BTW) is manufactured.

(We've hit the reply depth limit...)

> I can only comment on Carlos Castro, the link of yours at archivefreedom.org. Near as I can tell, he claims to have come up with a unified field theory. As far as I know, the arxiv filters out proofs of P=NP, the Riemann Hypothesis, unified field theory and similar things. So what?

Uh... so, it's scientists banning topics that contradict the widely held scientific consensus. That's what. Why is it just fine in physics but a sign of conspiracy in climate science?

> That's not the same as trying to get JMP shut down or editors ousted because it publishes papers on Bohmian mechanics.

If they published, without substantial peer review, one of the 'scientists' I linked to, it would be quite justifiable to get them shut down or their editors ousted, wouldn't it?

> With regards to angry muttering, there is far more than that. There is conclusive evidence of a conspiracy

That is conclusive evidence of one guy being a jackass.

But for the purposes of argument, let's assume he was indeed conspiring, rather than ranting. Who with? Who takes his orders and follows them?

All of the field of climate science? Riiiiiiiight. Are all of the many independent centers who are reproducing CRU research in on it too? And the many thousands of scientists with the expertise to uncover fraud are all willing participants?

All of this, established by this one email? Color me skeptical.

> The fact is, Steve McIntyre is not a crackpot.

That's your opinion of his plausibility, not a fact. (If you define "crackpot" as someone who repeatedly asserts views widely outside of the scientific consensus, without substantially altering said consensus, then he is, tautologically, a crackpot.)

> The CRU folks conspired to keep data from him in order to prevent him from making further corrections to Briffa's hockey stick.

I think it's a matter of the public record that he's been highly antagonistic to them. Are they "conspiring" or simply beginning to respond in kind? The latter is far more plausible to me.

> With regards your point c), the people who conspired to hide data now claim that other data was lost.

With regards to my (c), it's more that you were speculatively (or dishonestly) implying something that isn't true.

> With regards to a hacking war, I wasn't advocating it. I'm simply pointing out that the recent incident does give us evidence that the consensus (which is not a substitute for solid science, BTW) is manufactured.

That is just one interpretation of the emails.

An equally plausible interpretation is that these scientists are under intense political attack, taking it personally and getting angry, and beginning to unwisely respond in kind.

Who with? Who takes his orders and follows them? All of the field of climate science?

You are making a logical fallacy, namely the excluded middle. I never made the assertions you are mocking.

Are all of the many independent centers who are reproducing CRU research in on it too? And the many thousands of scientists with the expertise to uncover fraud are all willing participants?

Why don't you point me to these independent centers? Can you show me a case for contemporary warming which does not rely on CRU data?

Regarding Steve McIntyre, he corrected Mann's data. He altered the consensus, and (for a short time) had the best existing analysis. That, by definition, makes him not a crackpot.

Regarding antagonism, that's how science works. A good scientist will pick apart and search for flaws in any theory including his own. This weeds out the bad from the good.

If climate science is insufficiently antagonistic, that simply suggests their conclusions are more likely to be flawed.

An equally plausible interpretation is that these scientists are under intense political attack, taking it personally and getting angry, and beginning to unwisely respond in kind.

They are responding in kind within the scientific endeavor. That corrupts the scientific process, regardless of how emotionally reasonable their response is.

> You are making a logical fallacy, namely the excluded middle. I never made the assertions you are mocking.

You alleged a conspiracy. If there's just one person ranting, there is no conspiracy; and a conspiracy of small number of people wouldn't be sustainable in a larger honest community, particularly under the scrutiny they are under. It's not a fallacy in this case: there is precious little middle ground for you to stand on.

> Regarding Steve McIntyre, he corrected Mann's data. He altered the consensus, and (for a short time) had the best existing analysis. That, by definition, makes him not a crackpot.

That makes him not a crackpot by your opinion. By mine, he barely altered the consensus with a minor correction while wildly overstating the impact of his changes, spending far more time doing advocacy, public relations, muckraking and FUD than actual science.

> Regarding antagonism, that's how science works. A good scientist will pick apart and search for flaws in any theory including his own.

I did not mean "I don't agree with your hypothesis" antagonism, I mean "I think you're a fraud!" antagonism. I think it's an understandable if unfortunate human response to get pissy and withdraw from engagement; and I think that's far more plausible than any actual fraud or conspiracy.

> They are responding in kind within the scientific endeavor. That corrupts the scientific process, regardless of how emotionally reasonable their response is.

Ah, so their internal shop-talk, posturing, boasting, etc. is part of the scientific endeavour, is it?

What I've seen is that their "responding in kind" takes the form of advocacy and outreach (like RealClimate.org, which -- not coincidentally -- was the subject of several of the more damaging emails), not substantial alterations to the practice of their science.

(Sidebar: I won't say there might not be substantial alterations to the research itself; for a parallel example in a different field, there are evolutionary biologists who are now choosing research topics specifically to provide more compelling evidence for evolution, rather than explore the unknowns of evolution, due to the rise of the "intelligent design" debate here in the US.

However, the pattern there is to become -more conservative- in spending time and effort in making the obvious even more obvious, not more in becoming more alarmist or speculative.)

I alleged a conspiracy between a few people at the CRU to hide data, oust critics from professional societies and have journals shut down. Is that even in dispute?

Can you delete any emails you may have had with Keith re AR4? Keith will do likewise...Can you also email Gene and get him to do the same?...We will be getting Caspar to do likewise.

How do you interpret this as anything other than a conspiracy (between Jones, Keith, Gene and Caspar, not the entire climate science community, just to be 100% clear) to delete emails?

Rather than arguing against my simple, well supported conspiracy argument, you instead implied that I argued for some vast and implausible conspiracy. Why not just go for broke and act as if I claimed Phil Jones was behind 9/11?

Regarding McIntyre, whether or not he made a major improvement to the existing record or a minor incremental one, the point is that he is correct. A crackpot is a person with wild and incorrect theories completely unsupported by evidence.

Ah, so their internal shop-talk, posturing, boasting, etc. is part of the scientific endeavour, is it?

Ousting dissenters from professional societies certainly is, as is trying to get journals shut down.

Concerning antagonism, "I think you're a fraud" is the best kind to have. I wish it were more common, but it's too time consuming to use on an everyday basis.

> Rather than arguing against my simple, well supported conspiracy argument, you instead implied that I argued for some vast and implausible conspiracy. Why not just go for broke and act as if I claimed Phil Jones was behind 9/11?

If you're claiming or implying, as you have been until now, that 'conspiracies' are the only reason that the current scientific consensus is still standing, then it has to be a vast conspiracy, no?

> Regarding McIntyre, whether or not he made a major improvement to the existing record or a minor incremental one, the point is that he is correct. A crackpot is a person with wild and incorrect theories completely unsupported by evidence.

Even a stopped clock is right twice a day. We can argue back and forth as to whether he's a crackpot or not but all we'd be doing is arguing about the definition of the word. (And it would remain a matter of opinion and not fact!)

It is, however, true that his more substantial allegations practically require a widespread conspiracy, or something similarly dramatic, within the current consensus in order to be correct.

On the broad level of course its true. The rise of global warming alarmism over the past three decades has enormously increased the level of funding for research, and turned the original scientists into kingpins who get to play on the world stage. This doesn't necessarily mean that they are biased by the money and power, but it does mean you should treat grant funded scientists with the same level of skepticism that you would treat Exxon funded scientists (that is, if there actually were Exxon funded scientists, unfortunately, the defense has decided to not show up the case).

The typical modus operandi is that every time a "climate scientist" finds a shred of evidence that confirms their view, they run to the NYTimes with dire warnings. But every time their is evidence going the other way, they either hide it ( http://www.eastangliaemails.com/emails.php?eid=345&filen... ) or lie about it ( http://www.powerlineblog.com/archives/2009/11/024995.php ). Thus the layman reading only the NYTimes keeps hearing that "recent measurements have been worse than predicted", despite that there are dozens of other data points going the other way.

Further, let's assume the funding relationship is true. Are all these scientists that dishonest? Wouldn't honest ones be reigning in their colleagues? Why haven't we seen this?

First, they are not completely dishonest. It is not a conspiracy. It is a scientific theocracy. They are true believers. But if history shows anything, priesthoods are very common, even among smart, well meaning people.

And yes, there are dissenters. For instance, a bunch of scientists criticized the IPCC reports saying, "Finally, while the IPCC enlists many expert reviewers, no indication is given as to whether they disagreed with some or all of the material they reviewed. In previous IPCC reports many expert reviewers have lodged serious objections only to find that, while their objections are ignored, they are acknowledged in the final document, giving the impression that they endorsed the views expressed therein." and produced their own summary report. Read it here: http://www.uoguelph.ca/~rmckitri/research/ispm.html

But because the priesthood controls the peer review and grant process, most of the dissenters are either tenured professors like Richard Lindzen, or completely independent people like Steve McIntyre.

Occaissionally, you will get someone on the inside to write something very brave: "But I am also aware that in this thick atmosphere -and I am not speaking of greenhouse gases now- editors, reviewers and authors of alternative studies, analysis, interpretations,even based on the same data we have at our disposal, have been bullied and subtly blackmailed. In this atmosphere, Ph D students are often tempted to tweak their data so as to fit the 'politically correct picture'. Some, or many issues, about climate change are still not well known. Policy makers should be aware of the attempts to hide these uncertainties under a unified picture. I had the 'pleasure' to experience all this in my area of research." http://74.125.113.132/search?q=cache:ww6XwH5p28wJ:coast.gkss...

They're expected to present and defend their claims in the court of public opinion and in the realm of politics; why are you shocked and appalled that they are doing so?

Well, it's not supposed to be their job. Their job is to present a full and balanced view, it's the job of policy makers to make the policy.

But of course that's not the case. My real problem is that newspapers and people like you still treat these scientists as unbiased experts, rather than as expert witnesses for the prosecution. Listen to what they say, but don't treat the fact that there is a consensus as evidence that they are right. Of course all the expert witnesses for the prosecution are going to be of the same mind.

Sure. So does astrophysics, evolutionary biology, ... How is the situation different from any other deep speciality?

The advantage other sciences have is that their theories can be proven or disproven on a short time scale. Either your rocket works or it doesn't. Your medicine either kills people or saves them. Thus you have independent verification and falsifiability.

The area of sciences that are not subject to these forces tend to diverge from reality very quickly, see: macroeconomics, string theory, most nutrition research, etc.

So all we hear about is a "consensus" when that consensus is really just a self-replicating group of believers.

There is no shortcut other than to either look closely at the science yourself, or else find individuals that you really trust, and follow their opinion.

I haven't looked deeply into the actual science. But the amount I have looked into it raises an enormous number of questions. For a layman's summary of the main issues with the science, read the report above, and read these:

http://commentlog.org/bid/4409/The-Case-for-Global-Warming-S...

http://bishophill.squarespace.com/blog/2008/8/11/caspar-and-...

http://www.climateaudit.org/?p=1579

http://www.sciencebits.com/FittingElephants

> Thus the layman reading only the NYTimes keeps hearing that "recent measurements have been worse than predicted", despite that there are dozens of other data points going the other way.

And a layperson reading the WSJ hears the exact opposite. That's politics & demagogues, not climate scientists (in either case).

As to your links and quotes alleging blacklisting:

1) Blacklisting happens in science. Nobody has the time to respond to crackpots. This is acceptable when it's new-age physics kooks. It's acceptable when they are intelligent design advocates.

2) I can produce a list of links of people making the same accusations of suppression about evolutionary biologists; or about theoretical physicists. Why should climate 'outsiders' carry extra weight, when they have contributed correspondingly little to the field of climate science?

> My real problem is that newspapers and people like you still treat these scientists as unbiased experts, rather than as expert witnesses for the prosecution.

See, my problem is that newspapers include unqualified outsiders as 'balance' and treat them as having equal expertise, when the fact is that they are often completely unversed in climate science! (For example, the many economists treated as climate experts.)

Are they unbiased? No, although I would say their biases have a lot of individual variation.

Are they experts? Hell, yes. Do you really believe they aren't gaining deeper understanding of our climate through their work?

> The advantage other sciences have is that their theories can be proven or disproven on a short time scale. Either your rocket works or it doesn't. Your medicine either kills people or saves them. Thus you have independent verification and falsifiability.

This isn't even close to true. On a philosophical level, theories can never be proven (only dis-proven). On a practical level, there are huge amounts of established, consensus, science that cannot be tested on a short time scale. (Examples range from the big bang to human evolution; even for smaller time scales "short" is relatvie -- e.g. it's taken nearly a century for us to be able to test huge portions of Einstein's predictions.)

Further, there is a lot of independent corroboration (if not outright verification) in climate science. Do you think policy makers would be giving them the time of day if there weren't?

> So all we hear about is a "consensus" when that consensus is really just a self-replicating group of believers.

Of course a consensus is a group of people who share a belief; that's the #@$@#ing definition! And if they're "self replicating," maybe that would be because their arguments are compelling?

You're trying to have things both ways: There is no conspiracy, only a 'priesthood'. Yet your accusations here are that the religion is maintained by a conspiracy!

> For a layman's summary of the main issues with the science

Shall I send you convincing-to-a-layperson links about "irreducible complexity" as the main issue with evolution?

And a layperson reading the WSJ hears the exact opposite. That's politics & demagogues, not climate scientists (in either case).

That is true. I personally rarely read the WSJ or NYTimes because I don't find either effective sources of truth.

But the links I sent you come from much better sources, in my opinion, than either the WSJ or NYTimes. Did you read any of the links? I mean, you seemed to agree before when I said that the climate scientists were like lawyers for the prosecution. So have you read the case for the defense ( a case that is written by people much smarter than WSJ writers)? For example, why specifically so you think Nir Shaviv is wrong? http://www.sciencebits.com/FittingElephants

>But the links I sent you come from much better sources, in my opinion, than either the WSJ or NYTimes.

... For the link you repeated... blogs about politically charged topics are 'better source' than the NYT or WSJ? ;) I'd say they're equally horrible sources.

Why do I specifically think Nir Shaviv is wrong? I don't know that he is, but I know that he's failed to change the consensus, while his hypotheses have indeed been taken into account. (cite: a google search or two.)

So we're once again down to (a) a widespread conspiracy to hide the truth and force predetermined conclusions; or (b) an honest but imperfect process gradually iterating towards accuracy.

(N.b. If you accept his hypotheses about cosmic ray activity / solar warming, you also accept that warming is in fact happening (just not anthropogenic); if you find him convincing then by rights you ought to find McIntyre et al less so; yet everyone including you always seems to excerpt them side-by-side in some great litany of throwing it all at the wall and seeing what sticks...)

(Note: hit 'reply' by accident early, so this is an edit.)

blogs about politically charged topics are 'better source' than the NYT or WSJ? ;) I'd say they're equally horrible sources.

Blogs in general have a much worse average quality than the NYT, but a much higher peak quality. The links I sent you are among the best I have come across at making a clear case for the defense.

I don't in particular buy Shaviv's point about cosmic rays, but rather I agree with him on this point: "The fact that GCMs can fit the temperature increase is totally meaningless. Why? Because the allowed parameter space is so large climatologists could fit anything from a small temperature decrease (as was claimed in the 70's, before the imminent ice-age) to double or triple the observed 20th century increase."

Again, I have not studied climatalogy in depth. But I have studied economics in depth. I find the statistics and models published in peer reviewed journals and used by the central banks to be atrocious. I can read a very technical article and pull my hair out looking at the flaws. And lo, the predictions these models make invariably turn out to be crap. Yet these errors persist because the selection process in economics academia does not reward truth seeking.

I see the same pattern in climate science. There is a selection process that rewards conformity, not truth seeking. And a cursory look at the models and statistics shows that the scientists are indeed "fitting elephants". When you look at the raw unadjusted data, there is simply not much room for alarm. So until I can actually look at the data myself, or I can find experts that I actually do trust to give me an honest take, my default stance is skepticism.

(a) a widespread conspiracy to hide the truth and force predetermined conclusions; or (b) an honest but imperfect process gradually iterating towards accuracy.

or c) a priesthood. There is a big difference between a conspiracy and a priesthood. Priesthoods are very, very common in history, and they result in very widespread consensuses that turn out to be very wrong. See: The Catholic Church.

> "The fact that GCMs can fit the temperature increase is totally meaningless. Why? Because the allowed parameter space is so large climatologists could fit anything from a small temperature decrease (as was claimed in the 70's, before the imminent ice-age) to double or triple the observed 20th century increase."

Isn't this simply an assertion of fraud, made in a less direct manner?

> or c) a priesthood. There is a big difference between a conspiracy and a priesthood. Priesthoods are very, very common in history, and they result in very widespread consensuses that turn out to be very wrong. See: The Catholic Church.

The Catholic Church nicely matches my (a) whether you call it a conspiracy or a priesthood, and I'm sorry, but I don't see it.

If one of these outsiders had a truly viable alternative hypothesis, and could back it up, there is plenty of room for them to be heard.

> When you look at the raw unadjusted data, there is simply not much room for alarm.

... says one professed layperson to another.

Personally I find it almost inconceivable that, given how extensively we intentionally and unintentionally alter our local environments, we wouldn't be making corresponding impacts to the global environment. Argument-by-intuition, sure, but it's just as valid as yours!

Being a scientist and looking at the code that is written daily (and aware of the bugs that are found), I certainly agree that there is horribly written code going around. Most scientists aren't encouraged to learn about coding, it's something you do to get the job done. But like you say, that doesn't mean it produces incorrect output. Hopefully there are enough cross checks that mistakes are caught before publication.

I think you miss the fact that to figure out what scientific code does, you also have to know the science. Yeah, you can look for dumb bugs like the one they show in that clip, but you have no way of knowing whether calculations are correct without knowing what's going on.

I found it disingenuous how they flashed a line of code that says "fudge factor" as if that makes the code incorrect. Unless you know what that fudge factor is for, you can't make that judgement.

especially when it's numerical and certain approximations make scientific sense while others don't.

Climatologists should articulate what adjustments are necessary and why. Popularizers can then make it accessible to laypersons if need be - there is certainly enough interest.

The whole point is that up until now we've been trusting the details to climatologists, and that trust has been broken.

Layer on top the stakes, and also consider that this issue really lies at the intersection of many fields: climatology, computer science, economic, politics; opening this up is the only sensible next step.

Oh, I totally agree that it all should be public. I'm just saying that the OP's take that "let's get some professional programmers in there and clean it up" won't work unless they work in close collaboration with the scientists themselves.

While I agree with your comment I think it should be noted that it is a lot easier to reproduce results than to make original research. It would probably not be viable to come up with new climate models, etc. but it might be viable to verify and eventually correct the current models and algorithms. After all the science has been done, and is known, so checking assumptions behind the code should be doable.

I'd like to start a formal group to bring back the glaciers and arctic ice cap.

The neat thing about being a professional programmer is that you are always working in somebody else's problem domain.

So I wouldn't take the naysayers too seriously. If you can't peer review climate code, you can't peer review financial code, or medical code, or astronautics code.

Sure, you'll need problem domain expertise. But that's a given with any kind of programming. Climate science is no different than any other kind of work programmers do.

I think it's a great idea -- if you can pull it off. But that's a really big "if".

I'm definitely interested, not sure about the time.

Reply on news.ycombinator.com