...Why scientific programming does not compute

tullio
tullio
Joined: 22 Jan 05
Posts: 1,994
Credit: 32,281,599
RAC: 514

Glaciers are eternal, girls

Glaciers are eternal, girls are not. Nobel prizes are given more for political reasons than scientific merit.
Tullio

Michael Milan
Michael Milan
Joined: 3 Nov 05
Posts: 10
Credit: 1,102,562
RAC: 0

All very depressing indeed,

All very depressing indeed, but one thing that must be kept in mind is that - if the history of modern science is anything to go by - then the truth will eventually out, and science will self-correct from any errors of shoddy methodology/programming, whether in 10, 50, or 100 years.

Michael Milan
Michael Milan
Joined: 3 Nov 05
Posts: 10
Credit: 1,102,562
RAC: 0

RE: In my view a better

Quote:
In my view a better question is - how do you define 'scientific' behaviours? What deserves the valid use of that term as a description of areas of inquiry? Certainly climatology generally has taken a hammering in this regard. The CRU response to a divergence in the trends between dendritic ( tree ring ) and direct temperature data ought to have been to investigate the underlying physical reasons for the disparity. NOT to have chosen one set in preference to the other without good basis.

But the thing is, I believe they did have a good basis for choosing one set of data over the other. A very good reason indeed.

Direct measurements of surface temperature are precise, and the methodology is more or less incontrovertible. On the other hand, dendroclimatology is a far less accurate way of determining past temperature trends.

If the tree rings differ from modern, accurate, direct surface temperature measurements, you don't throw the baby out with the bath water.

The question you should be asking is: why should you NOT choose the far more reliable, consistent, and widely corroborated direct temperature data in preference to the tree ring data without good basis?

Quote:
By analogy if you had a particle accelerator and found that half the particles went one way while the other half the other, then you wouldn't choose one set and suppress the other data, but sort out why you appear to have two particle types. The first ( suppressive ) technique makes the data follow the conclusion, whereas the second is more honest - specifically you claim no conclusions until some can be made ( if at all ). Further data collection lights the way that your thinking should go.

That's not a good analogy in my opinion. A better one is this:

Say you have two particle accelerators firing a beam of protons.
One accelerator has had years of service, has been tested thousands of times and is known to give good results.
The other accelerator is very experimental and hasn't been tested much.
Now suppose the beam of protons go one way in the older accelerator as expected, but go in a different, unexpected direction in the newer, experimental accelerator.

Now, you don't sit on the results while you try to figure out whether you've discovered a fundamental flaw in the laws of physics - you simply assume that the results of the older, more reliable accelerator is correct, and then try to work out what went wrong in the design of the newer, experimental accelerator.

It's the same thing with direct temp data vs. tree ring data. If the trends diverge, you pick the more reliable data set, unless you have a damn good reason not to.

mikey
mikey
Joined: 22 Jan 05
Posts: 5,466
Credit: 531,101,193
RAC: 171,455

RE: RE: In my view a

Quote:
Quote:
In my view a better question is - how do you define 'scientific' behaviours? What deserves the valid use of that term as a description of areas of inquiry? Certainly climatology generally has taken a hammering in this regard. The CRU response to a divergence in the trends between dendritic ( tree ring ) and direct temperature data ought to have been to investigate the underlying physical reasons for the disparity. NOT to have chosen one set in preference to the other without good basis.

But the thing is, I believe they did have a good basis for choosing one set of data over the other. A very good reason indeed.

Direct measurements of surface temperature are precise, and the methodology is more or less incontrovertible. On the other hand, dendroclimatology is a far less accurate way of determining past temperature trends.

If the tree rings differ from modern, accurate, direct surface temperature measurements, you don't throw the baby out with the bath water.

The question you should be asking is: why should you NOT choose the far more reliable, consistent, and widely corroborated direct temperature data in preference to the tree ring data without good basis?

Quote:
By analogy if you had a particle accelerator and found that half the particles went one way while the other half the other, then you wouldn't choose one set and suppress the other data, but sort out why you appear to have two particle types. The first ( suppressive ) technique makes the data follow the conclusion, whereas the second is more honest - specifically you claim no conclusions until some can be made ( if at all ). Further data collection lights the way that your thinking should go.

That's not a good analogy in my opinion. A better one is this:

Say you have two particle accelerators firing a beam of protons.
One accelerator has had years of service, has been tested thousands of times and is known to give good results.
The other accelerator is very experimental and hasn't been tested much.
Now suppose the beam of protons go one way in the older accelerator as expected, but go in a different, unexpected direction in the newer, experimental accelerator.

Now, you don't sit on the results while you try to figure out whether you've discovered a fundamental flaw in the laws of physics - you simply assume that the results of the older, more reliable accelerator is correct, and then try to work out what went wrong in the design of the newer, experimental accelerator.

It's the same thing with direct temp data vs. tree ring data. If the trends diverge, you pick the more reliable data set, unless you have a damn good reason not to.

But aren't we then back to deciding what is right and what is wrong instead of letting the evidence/research tell us which is the right one? In your accelerator example the old one could have been wrong all along and the new one is FINALLY showing us the right way. If we don't look at alternative results then only the results that show what we want them too will be included, which is what the climate folks did! Alternative results need to be explained so that when someone else does the same research and finds different results they are not claiming we are frauds, as happened in the climate research. To just ignore data is wrong and akin to a cop walking up to a crime scene and arresting the person with the gun because there is a dead person lying on the ground! I am not saying the person with the gun did not shoot the person on the ground, but until some evidence is collected it could be a simple matter of someone picking up the gun to keep it from getting stolen or worse used again! Evidence is evidence, results are results, if we don't consider and try to explain either why are we doing what we do? Because we are no good to our Science or our job!!

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6,075
Credit: 116,533,118
RAC: 48,831

RE: if the tree rings

Quote:
if the tree rings differ from modern, accurate, direct surface temperature measurements


I think you're missing the point, and your wording indicates that you have some emotion which is over-riding critical analysis. Your annoyed aren't you? You want something to be true but there's a problem sooooo .... maybe if you say it three times it'll become true? But I'm discussing science not advocacy .... ;-)

I'm not discussing what the true trend is, nor what we wish to be true. We aren't climatologists. How could we possibly know or second guess what is not within our expertise? Scientific method is not "I assert something. No-one has contradicted me. Therefore it must be true".

Alas you're evidently not aware of the circularity within your own thinking eg.

Quote:
you simply assume that the results of the older, more reliable accelerator is correct, and then try to work out what went wrong in the design of the newer, experimental accelerator.


as it's assumption which is precisely the problem. This is a higher level matter that just examining accelerators, you have to have confidence in your intellectual method. Otherwise one 'begs the question' and you evidently haven't noticed that you've done exactly that. You're prepared to conclude the newer data is wrong because it is newer, and by construction eliminated any examination of the converse case. That's what 'simply assume' implies, you cut off further analysis toward a contrary case. With science you do the measurement with respect to reality to eliminate any assumption, and specifically treat all competing hypotheses in advance as equally worthy of critical attention.

Otherwise you will, not surprisingly, conclude precisely what you expected to find ( perhaps while even not realising that you implicitly chose that ). By thinking in advance that the older machine is more reliable ( an idea you might subtly self reinforce by using that word ), then you will conclude the new machine must be wrong ( which is labeled as the 'experimental' one, thus the wording prejudges it's performance ). Not because it necessarily is in reality, but because you implicitly left that as the only examinable hypothesis when you make the 'simply assume' step. Did you check the old machine for a ferret that died in it overnite? Maybe the new machine is not measuring the same thing you think the old machine is/was. There could be a blizzard of reasons for some disparity, which is why we let the data lead us so we don't suffer the failure of not imagining the right answer. So while you certainly ought check the new machine for flaws, don't forget to do the 'reliable' older one to. After all the space shuttle was pretty 'reliable' for nearly 100 launches wasn't it? But assumption on the behaviour of the O-rings killed seven astronauts, and another seven with falling tiles. NASA simply didn't learn their assumption problem, but there was no shortage of data that had been staring them in the face all along ......

The CRU problem was their own ( finally and painfully ) self admitted deliberate obfuscation of their methodology, data selection, and a prolonged adversarial campaign to prevent critical or contrary examination ( including several court cases ). They just didn't possess a methodology which allowed a bland unbiased treatment of the data sets. So I'm frankly not surprised they wound up where they did, but it was all their own work ( they ensured that ) so I have little sympathy.

The mathematicians have a terrific method to manage error on the road to their own 'truths' : they reveal all and allow, even encourage, full and frank detailed analysis of any and all hypotheses. Andrew Wiles admitted he took a terrible chance by working by himself for years ( on Fermat's Last Theorem ) as he had no one to check his work. He did make a few mistakes, but they ironed out OK after about a year of extra work.

Cheers, Mike.

( edit ) Can you see the problem that by not examining and understanding the disparity you then don't actually know if your basis for choice was good? One doesn't know what one doesn't know, so by avoiding direct physical knowledge and investigation then you diverge from understanding the real world when you choose to go that way. One risks ending up with a (poor) proxy form of the world - within one's own mind - a model that by virtue of it's construction may barely resemble the actual one ( and one won't even know how far off one is ).

I have made this letter longer than usual because I lack the time to make it shorter. Blaise Pascal

Michael Milan
Michael Milan
Joined: 3 Nov 05
Posts: 10
Credit: 1,102,562
RAC: 0

RE: But aren't we then back

Quote:
But aren't we then back to deciding what is right and what is wrong instead of letting the evidence/research tell us which is the right one? In your accelerator example the old one could have been wrong all along and the new one is FINALLY showing us the right way.

You're absolutely right, and I agree completely.

However, you need to remember that old saying: "Extraordinary claims require extraordinary evidence".

Going back to the two accelerators analogy; the reason why we choose to accept the results of the older accelerator is because of the huge body of knowledge we have concerning the behaviour of protons, built up over many years of consistent results, consistent theories, independent corroborating experiments, etc.
To put all of that vast body of knowledge into question, or even throw it all out of the window, because of a contradictory result from an experimental, not very well-tested accelerator requires a very very good reason why. Until a reason can be found, it should be the new accelerator result that should be put under further investigation alone, not the old accelerator. Extraordinary claims require extraordinary evidence.

Now, with the tree ring data, they indicate that the Earth had stopped warming since the 1950s. Why should we choose surface temp data in preference to this?

Because if we don't, then we have to explain why, if the Earth hasn't been warming since the 1950s, thermometers in locations all over the globe are still showing a consistent increasing trend in mean surface temperature for 60 years.
We have to explain, if the Earth hasn't been warming since the 1950s, why sea levels are rising.
We have to explain, if the Earth hasn't been warming since the 1950s, why glaciers are disappearing.
We have to explain, if the Earth hasn't been warming since the 1950s, why the permafrost is melting.
We have to explain, if the Earth hasn't been warming since the 1950s, how the concentration of atmospheric carbon dioxide, a greenhouse gas, can increase from 280 ppm in pre-industrial times to its current level of 370 ppm without having any effect on temperature whatsoever.

That's a lot of stuff to explain! Not only that, but all these observations are consistent with what we would expect if global warming is real. If, IF, we can find a way to explain all these observations in an alternative theory that also fits with the tree ring data, THEN we can be confident in questioning the direct surface temperature data, and giving the tree ring data an equal footing.

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6,075
Credit: 116,533,118
RAC: 48,831

RE: RE: But aren't we

Quote:
Quote:
But aren't we then back to deciding what is right and what is wrong instead of letting the evidence/research tell us which is the right one? In your accelerator example the old one could have been wrong all along and the new one is FINALLY showing us the right way.

You're absolutely right, and I agree completely.

However, you need to remember that old saying: "Extraordinary claims require extraordinary evidence".

Going back to the two accelerators analogy; the reason why we choose to accept the results of the older accelerator is because of the huge body of knowledge we have concerning the behaviour of protons, built up over many years of consistent results, consistent theories, independent corroborating experiments, etc.
To put all of that vast body of knowledge into question, or even throw it all out of the window, because of a contradictory result from an experimental, not very well-tested accelerator requires a very very good reason why. Until a reason can be found, it should be the new accelerator result that should be put under further investigation alone, not the old accelerator. Extraordinary claims require extraordinary evidence.


Oh, Michael. See the Michelson-Morley experiment for why that approach readily falls over. We don't have protons, we have an ( intellectual ) proton model. Your approach clearly aborts any possible refinement of that, even before any evidence has even been emitted from some new device. You are saying that a surprising finding must be wrong because it upsets your older model. 'Extraordinary' is a pre-judgment upon significance, what you expect isn't the issue - what the measurements actually indicate about the real world is. Heck if Penzias and Wilson had left their receiver assuming it just contained bird poo, they'd have missed out on the CMB - and they went there to calibrate radio signals bouncing off low earth orbit satellite 'balloons'!

Quote:

Now, with the tree ring data, they indicate that the Earth had stopped warming since the 1950s. Why should we choose surface temp data in preference to this?

Because if we don't, then we have to explain why, if the Earth hasn't been warming since the 1950s, thermometers in locations all over the globe are still showing a consistent increasing trend in mean surface temperature for 60 years.
We have to explain, if the Earth hasn't been warming since the 1950s, why sea levels are rising.
We have to explain, if the Earth hasn't been warming since the 1950s, why glaciers are disappearing.
We have to explain, if the Earth hasn't been warming since the 1950s, why the permafrost is melting.
We have to explain, if the Earth hasn't been warming since the 1950s, how the concentration of atmospheric carbon dioxide, a greenhouse gas, can increase from 280 ppm in pre-industrial times to its current level of 370 ppm without having any effect on temperature whatsoever.

That's a lot of stuff to explain! Not only that, but all these observations are consistent with what we would expect if global warming is real. If, IF, we can find a way to explain all these observations in an alternative theory that also fits with the tree ring data, THEN we can be confident in questioning the direct surface temperature data, and giving the tree ring data an equal footing.


Such are arguments of personal incredulity ( I just can't think of why that should be, so it musn't be? ). This is why we don't vet the data, and examine all hypotheses. You're judging the veracity of data by what you say it implies, thus reversing/mixing the roles of measurement, expectation and deduction. I don't think you have even realised you are doing that.

I'll leave you to your advocacy, Michael ... :-)

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter. Blaise Pascal

Michael Milan
Michael Milan
Joined: 3 Nov 05
Posts: 10
Credit: 1,102,562
RAC: 0

RE: I think you're missing

Quote:
I think you're missing the point, and your wording indicates that you have some emotion which is over-riding critical analysis. Your annoyed aren't you? You want something to be true but there's a problem sooooo .... maybe if you say it three times it'll become true? But I'm discussing science not advocacy .... ;-)

I guess I am being pretty emotional. :-P

Ok, I freely admit to being biased towards the global warming theory because I can't stand conspiracy theories. The stuff that some of the skeptics say sound eerily similar to how creationists say that secular scientists are taking part in a global, concerted effort to distort evidence in fossil records, radiometric dating, etc, in order to push evolution and a global atheistic agenda.

Sounds like how climate scientists are supposedly taking part in a global, concerted effort to distort evidence about global warming in order to secure more funding from government or push a global green/communist agenda.

Yes, just because you're paranoid, does not mean they're not out to get you, but I just have no patience for conspiracy theories.

So when climate scientists say that direct surface temperature measurements are sound and accurate and show an increasing trend, I put my trust in them in the same way I put my trust in planetary scientists that say the Earth is 4.5 billion years old.

Quote:
as it's assumption which is precisely the problem. This is a higher level matter that just examining accelerators, you have to have confidence in your intellectual method. Otherwise one 'begs the question' and you evidently haven't noticed that you've done exactly that. You're prepared to conclude the newer data is wrong because it is newer, and by construction eliminated any examination of the converse case. That's what 'simply assume' implies, you cut off further analysis toward a contrary case. With science you do the measurement with respect to reality to eliminate any assumption, and specifically treat all competing hypotheses in advance as equally worthy of critical attention.

Yes, but aren't assumptions ok to make within reason?. If you took the stance of not making assumptions about anything, then how could you get anything done? How would you calibrate an instrument if you can't assume that the older instrument you're calibrating it to still works?

And I never said that the result of the newer accelerator should be brushed under the rug and ignored. I said the course of action would be to investigate its design to see where it went wrong, if indeed it did. It should be investigated as an "open problem", tested and retested, perhaps with independent teams building another machine to the same specs and testing it themselves.

I don't believe that climate scientists are brushing the tree ring divergence problem under the rug either. I'm sure there are dendroclimatologists working away on the problem as I type. To think otherwise smacks of conspiracy theory, in my opinion.

Michael Milan
Michael Milan
Joined: 3 Nov 05
Posts: 10
Credit: 1,102,562
RAC: 0

RE: Such are arguments of

Quote:
Such are arguments of personal incredulity ( I just can't think of why that should be, so it musn't be? ). This is why we don't vet the data, and examine all hypotheses. You're judging the veracity of data by what you say it implies, thus reversing/mixing the roles of measurement, expectation and deduction. I don't think you have even realised you are doing that.

Personal incredulity, perhaps, but the point is that all these observations already have a cohesive, elegant, overarching theory to explain all of them: ie. anthropogenic global warming.

If skeptics can come up with satisfactory alternative explanations for all of these things, then I'll give it some thought, but to my knowledge they haven't (yet).

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6,075
Credit: 116,533,118
RAC: 48,831

Well I was only originally

Well I was only originally commenting on the CRU thing ( I hope you read my earlier hope that it was an 'outlier' ). Heaven forbid we should tar the entire field. Our climate is a serious issue, and I don't want the serious study of our civilisation's effects on that perverted by 'junk science'.

With the thrust forward into centre stage that climatology has 'suffered' we just don't know what to reasonably assume do we? I'm sure they do of course. But who in the field would be game to pop their hand up and give us the low down now? Alas we only seem to get derivative drivel by agenda pushers outside of the field and the Flip-Flop-Flannerys selling copy. There are now so many sociological connections to the topic - which isn't terrible per se, but a tad overwhelming and quite removed from the data, whatever it may be.

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter. Blaise Pascal

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.