pretty good rant

The two physics examples are pretty weak. (both retracted) I never heard of the BICEPS thing, and the faster than light neutrinos was pretty much thought by everyone to be a electronic glitch somewhere.

Reading between the lines are they trying to dis climate science?

I don't know the parent "paper" "First Things".

Oh I found this,

Will Happer knows lots about Optical Pumping. (He's helped us in the distant past.)

George H.

Physics is pretty good. It's a "hard" experimental science and is reasonably fad-resistant. Physisists seem to be unforgiving of error.

I sure hope so.

Medicine is interesting. Drugs seem to work and then they don't.

We probably have too many scientists, all struggling to publish. If 10 people do basically the same study, and one gets positive results, he gets to publish. And get a blurb on the science web sites.

Den mandag den 18. april 2016 kl. 23.49.56 UTC+2 skrev John Larkin:

afaiu at many universities, publishing a certain amount is either a requirement or tied to bonuses

-Lasse

The article's on the money, though overly kind to research judging by the p apers I've read.

There are problems at every stage, not just some. Most researchers simply lack the necessary reasoning ability to produce a l ogically valid paper. Results are easy to plausibly deniably skew in various ways. Given the stro ng incentive to do so it's inevitably common. Very common. Very very common . Confidence values are too often valueless since they fail to address all th e sources of errors. Many studies have conclusions tacked on that simply don't match the rest of the paper content. Journals get paid to publish, so are looking for content to print even when there's nothing genuinely worth printing. Some entire disciples are largely cobblers from the ground up, psychiatry a nd abnormal psychology especially.

Then we come to peer review, the main way faults are imagined to be picked up & detected. Last year I saw a shocking study directly related to what I do, but the reality is I have so many other things to do and didn't end up responding. The atricle has been accepted as truth now, and what it recomme nds is mostly done, much to the detriment of many.

The system is truly broken. We have to work with what we have, just take st udy findings with the pinch of salt they almost all deserve. We can at best work with a lot of very uncertain 'knowledge,' much of which is simply wro ng. Since double blind placebo controlled trials are so poor, uncontrolled trials & other data that lacks such controls also has useful validity, albe it with its own problems. Dismissing such data is unwise - as with all data one must look at the figures, arguments & conclusions and bear in mind the serious problems almost always present with them.

NT

I can only speak of the physics end, but there use to be small teaching colleges. Now all those physics faculty are expected to do original work, publish, get grants, etc.

Science has become a commodity, pages in print it's capitol. Perhaps it's reaching it's first inflation crisis... (I hope that's not too many mixed metaphors. :^)

George H.

John's ideas about climate science come from the Murdoch media. If he had t he attention span to slog through the American Institute of Physics web-sit e on the subject

he might become better-informed, but he he probably skipped too many physic s lectures at Tulane to be able to follow the actual science.

Drug trials are expensive, which means that they get paid for by the compan y that hopes to sell the drug. The medicos running the trials don't have th e same strong financial interest in the result, so mostly the results can b e trusted, but bad apples are bound to show up from time to time.

Most scientists know roughly what their colleagues are doing, and try and s teer their students into doing new stuff that other people aren't working o n.

Anybody who does new and interesting work gets attention. In a sense scienc e is an aspect of the entertainment industry, though they largely entertain other specialists in their field.

Technology spin-offs are an incidental benefit - they've given us our prese nt high-tech civilisation, and seem to deliver new goodies at an ever-incre asing rate, but this is built on a culture of curiosity about what's going on, rahter than some imperial scheme to conquer nature.

papers I've read.

It's rubbish - alarmist rubbish.

logically valid paper.

Anybody who has refereed papers for publication can confirm this.

rong incentive to do so it's inevitably common. Very common. Very very comm on.

And looked for very carefully. You've got to be pretty subtle about it to g et it past referees.

the sources of errors.

Confidence limits are technically based on the variance of the data being p ublished. Systematic errors don't create variance, and can be missed.

Independent studies tend to have different systematic errors, and if they g ive different results, everybody involved starts looking for the answer (wh ich thay will be able to publish).

of the paper content.

Anybody who has refereed papers for publication can confirm this. It does t end to prevent the paper getting published.

en there's nothing genuinely worth printing.

Journal with any kind of reputation get more papers than they can publish. The referring system winnows the papers down to stuff that won't damage the reputation of the journal (which is its biggest asset).

and abnormal psychology especially.

There are psychiatric procedures that work - cognitive therapy comes to min d. The human mind is remarkably complex - it comprises about 86 billion neu rones - so human psychology has quite a lot to get to grips with.

d up & detected. Last year I saw a shocking study directly related to what I do, but the reality is I have so many other things to do and didn't end u p responding. The article has been accepted as truth now, and what it recom mends is mostly done, much to the detriment of many.

It has been published, but probably not "accepted as truth". If you read mu ch of the published literature you get to realise that Sturgeon's Law appli es - 90% of everything is rubbish - and treat everything you read with cons iderable scepticism.

study findings with the pinch of salt they almost all deserve. We can at be st work with a lot of very uncertain 'knowledge,' much of which is simply w rong.

Much of which is misleading, rather than wrong. "Wrong" is relatively easy to prove, and you can get to publish a paper correcting a demonstrable erro r. "Misleading" rarely offers a chance to publish a correction.

Why do you think they are "poor"? They are expensive and hard to set up, an d convincing placebo's can be difficult to find (and even more difficult to get through ethics committees, but they do work.

validity, albeit with its own problems. Dismissing such data is unwise - as with all data one must look at the figures, arguments & conclusions and be ar in mind the serious problems almost always present with them.

So what? The fact that you always have to think hard about what others peop les data actually means doesn't invalidate the scientific method.

The scientific method isn't any kind of magic - it's just the way we've evo lved to get the best information we can out of the work we do.

Half-wits claiming that it's "broken" because it isn't perfect don't really deserve any attention.

Tenure, certainly. This has always been the case, though.

Publishing good papers and having other people cite them is a pretty good i ndication of scientific competence, and hard to fake.

Since universities are about teaching students, teaching competence ought t o come into it, but measuring it is more difficult, and people who are good at doing research, and communicating what they have found out, tend to be good at communicating basic ideas to students. There is a tendency to let g ood people concentrate on research and load the teaching onto the less stel lar performers, though that's betraying what universities are set up to do.

Sadly, you get what you measure, and the system needs a better measure of t eaching competence.

The key issue with the faster than light neutrinos is that the scientists involved in the measurement thought it was a glitch. They had spent a great deal of time and effort trying to find the glitch or experimental error, but failed to do so. Eventually they decided to publicise the work around the issue, primarily to get help and ideas that would help them find the error - and with a secondary purpose of noting that if there was no error, the others should try to replicate the result because it would be so revolutionary.

However, the mass media spread this as "scientists claim to have broken the light barrier" - because that is a far more exciting message than "scientists have an error in their equipment, but can't find it".

In science, it is normal to publish preliminary data, theories or results - spreading information and gathering ideas and opinions is vital in science, and publishing data or conclusions that later cannot be reproduced is par for the course. It is how science works - it is only after others have reproduced the same results that one can start to call it "scientific theory", "fact", or reliable information. However, the mass media likes to take the first publications and blow them out of proportion and out of context as a revolution or revelation, even though scientists know fine that they have only made step 1 on a long path that could easily be a dead end.

It is not "big science" that is broken (though there is always scope for improvement - the article makes some valid points). It is the media that is broken.

For a lot of information about this, with special emphasis on medical and pharmaceutical issues (since the crusader behind this is a medical doctor, and it's an area that affects us all a good deal more directly than neutrinos), I recommend this site and the "Bad Science" book:

I did - probably because I get to see Physics Today.

The problem with that was that the physicists doing the measurement got the result they were looking for, but didn't look hard enough for confounds - other effects that could have given them the same result.

Stephen J Gould's "The Mis-measure of Man"

is a full bottle on the tendency of researcher's to be less enthusiastic ab out looking for confounds when they've found the result they expect.

es

But he doesn't know much about anthropogenic global warming, and what he cl aims to know he seems to have got from the denialist song-sheet. There's a lively business in retired elderly experts with a scientific reputation in some field not all that closely related to climate selling that reputation to the denialist propaganda machine to give it a bit more credibility with those who can't be bother to read the rubbish being peddled.

Happer managed to write off greenhouse warming in a very few lines.

"The frightening warnings that alarmists offer about the effects of doublin g CO2 are based on computer models that assume that the direct warming effe ct of CO2 is multiplied by a large "feedback factor" from CO2-induced chang es in water vapor and clouds, which supposedly contribute much more to the greenhouse warming of the earth than CO2."

The feedback factor from water vapour isn't large, but it is significant, a nd there's nothing "supposed" about it. Water vapour is a greenhouse gas, a nd higher global temperatures put more of it in the lower levels of the atm osphere. 70% of the earth's surface is covered by ocean, and the water vapo ur content of the air above that ocean is determined by the vapour pressure of water, which rises as the surface temperature gets higher.

Cloud feedback doesn't really come into it - there doesn't seem to be all t hat much of it, and what there is goes both ways. Lumping it into the predi cted consequences of more CO2 in the atmosphere is decidedly disingenuous.

Increased cloud cover didn't save the planet from the last greenhouse gas e vent - the Paleocene-Eocene Thermal Maximum, some 56 million years ago whic h - if the isotope ratios are anything to go by - was driven by the dissoci ation of a lot of methane hydrate. We might be able to do that ourselves.

The day of Big Science is over.

The day of Weird Science has just begun.

One of the problems with some of the classes I took in the cognitive science department of the college I went to is that they DID have people good at research teaching classes.

The problem was, I got the sense that some of these professors viewed their teaching role as a burden, something they grudgingly did in their spare time because it was required of them to keep the grant money coming in.

They were some of the worst classes, because of the sink-or-swim attitude that these guys had towards didaction. To survive you pretty much had to come into the class knowing the material already, and they played real favorites - the classes basically turned into farming grounds for the professor which he used to scope out hot talent that might come in useful for his projects. Not much patience for everybody else.

Two of the guys in my freshman "Artificial Intelligence Programming Concepts" class ~15 years ago are now employed by the department in some capacity.

Cool site.

Not so. "Publish or perish" is a post-WW2 phenomenon. It has destroyed the humanities, because it isn't enough to be learned, you have to be original. Nobody gets tenure by agreeing with his predecessor, so inevitably we go off into the swamps of fashion and absurdity. It's as Deep Thought said in "The Hitchhiker's Guide to the Galaxy:

========== "And it occurs to me that running a programme like this is bound to create an enormous amount of popular publicity for the whole area of philosophy in general. Everyone's going to have their own theories about what answer I'm eventually to come up with, and who better to capitalize on that media market than you yourself? So long as you can keep disagreeing with each other violently enough and slagging each other off in the popular press, you can keep yourself on the gravy train for life. How does that sound?"

The two philosophers gaped at him.

"Bloody hell," said Majikthise, "now that is what I call thinking. Here Vroomfondel, why do we never think of things like that?"

"Dunno," said Vroomfondel in an awed whisper, "think our brains must be too highly trained Majikthise."

So saying, they turned on their heels and walked out of the door and into a lifestyle beyond their wildest dreams.

=========

Cheers

Phil Hobbs

he result they were looking for, but didn't look hard enough for confounds

- other effects that could have given them the same result.

about looking for confounds when they've found the result they expect.

ases

claims to know he seems to have got from the denialist song-sheet.

Hi Bill, I don't want to "do" the whole AGW thing. I like Will, One thing he certainly gets right is that climate science has become political.

"A major problem has been the co-opting of climate science by politics, amb ition, greed, and what seems to be a hereditary human need for a righteous cause. What better cause than saving the planet? Especially if one can get ample, secure funding at the same time? Huge amounts of money are available from governments and wealthy foundations for climate institutes and for cl imate-related research."

That's a bad thing for science. There should be more dissent, disagreement , lively debate. The fact that ~99% (or whatever the number is) of climate scienti sts agree with AGW tells me that dissent has been squashed in the name of political u nity.

George H.

There's a lively business in retired elderly experts with a scientific repu tation in some field not all that closely related to climate selling that r eputation to the denialist propaganda machine to give it a bit more credibi lity with those who can't be bother to read the rubbish being peddled.

ing CO2 are based on computer models that assume that the direct warming ef fect of CO2 is multiplied by a large "feedback factor" from CO2-induced cha nges in water vapor and clouds, which supposedly contribute much more to th e greenhouse warming of the earth than CO2."

and there's nothing "supposed" about it. Water vapour is a greenhouse gas, and higher global temperatures put more of it in the lower levels of the a tmosphere. 70% of the earth's surface is covered by ocean, and the water va pour content of the air above that ocean is determined by the vapour pressu re of water, which rises as the surface temperature gets higher.

that much of it, and what there is goes both ways. Lumping it into the pre dicted consequences of more CO2 in the atmosphere is decidedly disingenuous .

event - the Paleocene-Eocene Thermal Maximum, some 56 million years ago wh ich - if the isotope ratios are anything to go by - was driven by the disso ciation of a lot of methane hydrate. We might be able to do that ourselves .

"Big Science's conclusions about climate change don't agree with what we find attractive, therefore it's broken"

Fair enough. Perhaps I'm not as old as you are. ;-) That's close enough to "always" for me (yes, I was going to say "a long time" but didn't).

I remember my father bitching about it and it wasn't new for him. His career split WW2, though mostly on the post side (left OSU when the war broke out).

Tenure, itself, hasn't helped.

The joke about that in then movie Ghostbusters is when Ray Stanz turns to Peter Venkeman, when they're discussing starting a paranormal investigation business and says: "Look, you're an academic. You've never been out of the university! You don't know what it's like out there. _I've_ worked in the private sector. They expect...results."