Comments on: McKitrick mucks it up

By: liberal japonicus

liberal japonicus — Mon, 30 Aug 2004 04:22:51 +0000

John, shonky = sh*tty + wonky?? And thank god for Tim Lambert.

By: dave heasman

dave heasman — Sat, 28 Aug 2004 00:37:43 +0000

I was on the inside of the Y2000 software project. Mirror Group newspapers spent £170 000 for 4 experts to spend a year ensuring that adverts could be booked (in advance, obviously) over the year-change. We inspected programs, designed a coherent solution, changed and tested 800+ programs, drove it through user-testing and against an unchanged suite. Hard, fairly rewarding work. Almost certainly worth the money to the customer. All it took to make it cost-effective was for one competent person at the customer site hiring an expert contractor to scope the project and then hire more experts directly. If they’d used a major software house they’d have got inferior results and spent over a million. Ho hum, as the New Yorker has stopped saying.

By: Dem

Dem — Fri, 27 Aug 2004 23:40:24 +0000

joe silverman: the y2k example is an example of the familiar argument that the wasteful and mostly unnecessary steps we took to avoid disaster must have worked…because the disaster didn’t happen as predicted. I’ll agree with you on a lot of what you said, but let’s not pretend there wasn’t truly a severe problem to be solved with Y2K. Yes, you are correct that the lack of a disaster, by itself, doesn’t prove there ever was a problem. You are also right that most of the steps taken were unnecessary. Once the ball got rolling lots of people took advantage of Y2K to either sell stuff to unsuspecting people (thanks, MSFT), or (in the case of IT depts) use Y2K to justify getting stuff they wanted but couldn’t get approval of otherwise. However, I too was on the inside of Y2K, having worked at a major computer vendor, and I took away three lessons. First, there really was a serious problem. We found lots of things in our operating system code that would have taken the computers down, either on that day or soon thereafter. Second, that a lot of noise had to made by engineers before the leaders reacted (it wasn’t until the CEO realized he was personally liable if precautions weren’t taken that he took action). And third, that all that noise we made caused an over-reaction, causing huge overexpenditures.

By: joe silverman

joe silverman — Fri, 27 Aug 2004 19:49:54 +0000

the y2k example is an example of the familiar argument that the wasteful and mostly unnecessary steps we took to avoid disaster must have worked…because the disaster didn’t happen as predicted. as i was on the inside of that “disaster” i can attest to the fact that there were a lot of groups that found it useful to (1) predict disaster and (2) profit from a lot of unnecessary efforts to avoid it. i’m not on the inside of possible environmental “disasters” so i can’t comment intelligently on that subject…but i am cautious about all claims on almost all subjects.

By: Neil Sinhababu

Neil Sinhababu — Fri, 27 Aug 2004 07:26:30 +0000

I just want to affirm that Tim Lambert is awesome.

By: Ethesis

Ethesis — Fri, 27 Aug 2004 01:26:57 +0000

Good points on Crichton—it was interesting to discover he was as wrong on other points as he was on second hand smoke. Thanks!

By: Matt Weiner

Matt Weiner — Thu, 26 Aug 2004 22:24:50 +0000

It seems to me that if McKittrick is half-honest his work isn’t shoddier than Lott’s; Lott made up data. OTOH if McKittrick said there is no such thing as a Global Temperature. Temperature is a continuous field, not a scalar, and there is no physics to guide reducing this field to a scalar, by averaging or any other method then if I’m not mistaken he’s a buffoon. Can’t you integrate over the space on which the field takes values?

By: John Quiggin

John Quiggin — Thu, 26 Aug 2004 20:54:53 +0000

My guess is that the mistake was “half-honest”. That is, M&M were running lots of regressions, trying to find a specification that gave them the answer they wanted. In particular, I’ll bet they used absolute latitude before trying the cosine and found it highly significant. When they found that their cosine variable was insignificant, and the economic variables came out “right”, they didn’t look to hard at the diagnostic stats.

By: Thomas Palm

Thomas Palm — Thu, 26 Aug 2004 19:09:42 +0000

Crichton may have a lot he claims he can tell us, but that doesn’t mean he is correct. His statements about DDT are misleading at best.
DDT has never been banned for use to fight malaria. What caused the resurgence in malaria was lack of funding and increased resistance both to DDT, other pesticides and the most common medicines against malaria. The way DDT was used in agriculture and sprayed everywhere to get rid of annoying but harmless insects helped promote resistance. Researchers knew this from the start and urged that DDT should be reserved for fighting malaria so that it would remain effective as long as possible, but no one listened to them. If he wants to blame someone Crichton should blame the farmers.
In the same way Crichton claims that DDT is harmless. This isn’t really true, but it has a grain of truth. The major damage to birds come not from DDT but from DDE, which is formed when DDT breaks down in nature. Thus DDT is still to blame.

By: Dem

Dem — Thu, 26 Aug 2004 17:59:47 +0000

Crichton’s article was interesting, with some good points. But it was riddled with bad logic. 1) I liked the opening discussion about how the idea of Eden—the “idyllic natural environment before white man spoiled it”—is a myth. But in setting this up he’s setting up a straw man of the environmental movement as people who love nature but don’t know what nature really is. The most strident environmentalists I know all spend lots of time in wilderness experiencing nature. And not all nature is as hostile to humans as the Borneo example … consider John Muir setting up his cabin in Yosemite as a counter-example. 2) His comment about the predictors of overpopulation getting it wrong is equivalent to saying those who predicted Y2K computer problems got it wrong. He misses the key point that because of the predictions actions were taken to counter the impending disaster. In both cases, the responsible scientists predicted that “unless action was taken to change the trend” disaster would strike. In Y2K, billions were spent testing and upgrading software and hardware to prevent the disaster. For the population explosion, billions have been spent researching and distributing contraception, and researching and addressing the root causes of overpopulation. 3) His paragraph that starts talking about second hand smoke pretty much discredits the validity of the entire piece. He doesn’t give the references behind each of his startling claims, but if you have any knowledge of these areas you know he is cherry-picking data and ignoring the whole.

By: Sam

Sam — Thu, 26 Aug 2004 14:46:31 +0000

I must say that I’m only slightly surprised, but can easily believe that the mistake was an honest mistake. I trained in Econometrics, and worked for almost a year once on a paper with a major math error. I took the derivative of x^2 as x instead of 2x, and used the code-block with that error pervasively in my programming. The paper was reviewed by colleagues numerous times, and no one caught that error until the last review before I would have submitted it for publication. It’s not hard for me to believe that the error of inputting values in degrees when they should have been in radians was an honest mistake—one which I’m glad someone found.

By: praktike

praktike — Thu, 26 Aug 2004 13:37:52 +0000

See, Chrichton wasn’t calling for real science, though. He was calling for garbage like M&M to be accepted just because it bucked the consensus, which wasn’t exactly arrived at willy-nilly.

By: Ethesis

Ethesis — Thu, 26 Aug 2004 13:01:29 +0000

“Unless we’re right.” Might I ask, about what? Global cooling? Obviously significant steps need to be taken to continue towards strategic environmental goals. Crichton only makes the point that the current environment is driven by people who are religious about it rather than scientific, with his starkest example being the 20-30 million already dead from the anti-DDT campaign. Of course Crichton’s background in health care affects the way he feels about those deaths. I’ve talked with people who felt that they weren’t enough. I think the points are interesting, though I agree that he is wrong on some of his presumptions (which makes me question some of his conclusions). But he is really calling for real verification and real science, not religion in addressing real problems (assuming that Vernaculo sees the environmental problems as real rather than as religious props). Makes sense, the environmental problems we face are real. Real science (such as the large particle studies fueled by the Geneva Steel statistics showing that they are actually a threat to human health) seems to be called for. I don’t know about you, but I have children, and I want a better world for them.

By: vernaculo

vernaculo — Thu, 26 Aug 2004 10:38:34 +0000

Crichton- “The greatest challenge facing mankind is the challenge of distinguishing reality from fantasy, truth from propaganda.” Unless we’re right. Then the greatest challenge facing mankind would be violently catastrophic ecological disruption. During which being able to tell the true from the false would certainly still be vitally important. Just as it is when driving or playing golf.

By: Realist

Realist — Thu, 26 Aug 2004 03:29:51 +0000

[] Interesting article. BTW, he is quite wrong on details (such as second hand smoke) where there is clear, clean evidence. Note from JQ: I deleted a repost of an entire article by Michael Crichton here and replaced it with this link.