...Why scientific programming does not compute

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6534
Credit: 284706835
RAC: 110483

True scientific behaviour,

True scientific behaviour, which began around the time of The Renaissance, involves an 'arc of reasoning'.

Firstly one would examine some subset of the Universe, for Tycho Brahe it was those lights in the sky.

Then after study one would 'induce' a pattern on the data. It's a guess on what might be some common thread/cause/factors/whatever.

Then a general rule would be formulated to describe/summarise the pattern. But there is more ....

That general rule would then be applied to some other data set not yet considered, in order to determine what that rule would predict ought be in this other data - should the rule be true.

Finally a comparison would be made between the expectation deduced from the rule and that data.

It is ever so crucial that the rule be applied elsewhere, away from the data that was used to construct the rule. Simply re-applying back to the original data set is mere tautology, a kind of reverse gear that puts you back to the spot you started. At most it can say that you didn't make a logical error in reverting. Otherwise you wind up 'begging the question' as explained earlier.

Now in the CRU example, they failed in the set selection : by choices made they ensured their further analysis referred to a planet in a solarsystem. Just not this one. Their induction was probably right upon the selected data, but they only chose one pattern to induce leaving out other possible conclusions consistent with the set. Finally I don't see that they ever broadened to other data, or if they did whether a comparison ( b/w rule and data ) was performed, or if they did whether that fed back to the rule modelling ( because you can recurse with this arc, indeed one should ). The management of the tree ring issue heavily hints towards the 'nay' here. Their response was to not change the hypothesis and emerge with an updated rule, but to trim the data considered thus leaving the rule unchanged.

Now I am, again, going to sound like a real pain. But most of what I hear on this climate business is induced patterns. The full arc is not supplied, and what parts I have seen are badly broken. So what remains is advocacy, not science, and largely emitted by the spin producing wordsmith whizz-kids. The logical thread, if any, remains at pre-Renaissance level, which is why we wind up with the Medieval equivalents of portents of doom in the sky from comets, and she must be a witch because she weighs the same as a duck.

A more believable thread is much easier, and less work too, that being the various proponents/opponents/participants are merely preaching to their own converted and orating to secure their own place within some milieu that supports and nourishes them. So you get the Church of Climate Change with pastor Flip-Flop giving the latest sermon, and naturally with donations. Any scientific truth, as discussed above, is and has always been irrelevant. It's the scientific label which is sought, to obfuscate deeper agendas and intent. I mean what's 'clean coal' when it's at home? Like an old advertising ploy for Colgate toothpaste for instance : 'University tests prove that ....'

Other errors are more general than tangled/shattered inductive/deductive arcs. False dichotomy is frequently used - here's two options, I say it's not A thus it has to be B. This ain't proof by contradiction, but a contradictory proof! Another is intellectual laziness, hinted at with this 'extraordinary claims' line - it can't be right because I can't think of a reason, or it must be right because you haven't counterclaimed my assertion. But if one abrogates responsibility for un-thought-of patterns & rules ( by defining or marginalising them out of the discussion ) then one ought not complain when either better data or a better rule arrives. Or even worse : when plain fact contradicts. I not sure paranoia or conspiracy is required to follow the game, as mentioned mere stupidity and sheer greed are powerful enough cohorts already.

Cheers, Mike.

( edit ) BTW there will always be another data set to apply any general rule to. You just simply wait until tomorrow's measurements arrive! :-)

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

... and my other CPU is a Ryzen 5950X :-) Blaise Pascal

mikey
mikey
Joined: 22 Jan 05
Posts: 11888
Credit: 1828071059
RAC: 206164

RE: RE: It is kind of

Quote:
Quote:

It is kind of like Climate change...it IS going to happen, in fact it IS happening right now!! The only thing in question is what is the Earth getting, warmer or cooler. That of course leads to all the other questions...ie what caused it, how long will it last, etc, etc, etc!]/quote]

If you wish to try to argue by analogy, something that is impossible in science -- a thing is what it is not an analogy, then at least lets keep it straight. There is going to be a political response to the media advocacy of this magic global warming. That says nothing about the warming itself.

As for the exclamatory "is happening" I rebutted several things supposedly happening right now that are not in fact happening that anyone can tell by real world experience. As long as the "rising" sea level is never noticeable but always ten years in the future there is reason to work up a sweat about it.

Sorry I was not clear, "is happening" was meant to mean that today it is cold, tomorrow it is warmer. That has nothing to do with any long term variables, it is just that in the moment Climate Change "is happening". It was not meant to be Scientific but instead more of a tongue in cheek observation.

Quote:
It is also like my friend 'needing' a new pc, he wants one, has the money for one, and IS going to get one. But does he 'need' one, no he does not, but darnit he IS going to get one! That facts say and PROVE he does not utilize 1/4 of the the current pc's potential, but darnit it is 3 years old and he WANTS a new pc!! He has convinced himself that his email will travel thru the network faster on a new pc then his current pc, but he is mistaken, that is not his problem! His problem is that the program he uses is not fast, it is free and he sees ads instead of mail transfers! His other problem is he cannot brag about his pc anymore, he is not the guy with the latest and greatest pc anymore, so he 'needs' a new one!! Facts be darned, he WILL be getting a new pc shortly and CostCo, or Dell, will be a little richer for it.

Do your friend a favor and tell him about tigerdirect.com. I have no connection but as a satisfied customer. And with a little luck there might be a refurbished quad core AMD available for the price of a new dual core AMD. I have bought two like that in the last year. If he is on a tight budget tell him to look for a refurbed dual core for the price of a new single core.

I do know about Tiger and have bought from them many times and too am a satisfied customer, my friend is just stubborn and likes the 'double the manufacturers warranty' on all electronics that Costco gives. It also gets me 'out of the loop' for any repairs, he will just take it to them! Although for me repairing pc's is not a problem, I just WILL NOT work on a pc under warranty, that is silly and can be counter productive in the end.

Rod
Rod
Joined: 3 Jan 06
Posts: 4396
Credit: 811266
RAC: 0

Good Coding is just a tip of

Good Coding is just a tip of the iceberg.

Changing Software and Hardware in the Scientific Community

Preserving Science Choosing What Data to Discard

There are some who can live without wild things and some who cannot. - Aldo Leopold

Ver Greeneyes
Ver Greeneyes
Joined: 26 Mar 09
Posts: 140
Credit: 9562235
RAC: 0

My apologies as I have not

My apologies as I have not read through the entirety of this thread. But I would like to mention Bruno Latour here, who argues in his 'An attempt at a "Compositionist Manifesto"' that increasingly science is not about revealing some underlying truth about reality, but about ways of compositing data. The question then becomes not whether something is True or False, but whether the composition is good or bad - which is much more in line with the normative choices scientists are forced to make in climate predictions, life cycle analyses et cetera without undermining their value.

Latour uses (in his usual witty manner) the example of the 'climate gate scandal' to examine the modernist worldview that most, including scientists, believe in so firmly - and finds that it is inadequate when attempting integrate science into political decisions on policy. He further proposes that rather than speaking of 'matters of fact' we should consider the object of science to be 'matters of concern', which allows us to appreciate the value judgments that scientists must make rather than dismissing them as unscientific.

Note: I am by no means an expert on Latour; I came across his texts recently in the context of morality implicit in technology, and found them both entertaining and insightful. I'm curious what others in this topic think of his stance.

mikey
mikey
Joined: 22 Jan 05
Posts: 11888
Credit: 1828071059
RAC: 206164

RE: My apologies as I have

Quote:

My apologies as I have not read through the entirety of this thread. But I would like to mention Bruno Latour here, who argues in his 'An attempt at a "Compositionist Manifesto"' that increasingly science is not about revealing some underlying truth about reality, but about ways of compositing data. The question then becomes not whether something is True or False, but whether the composition is good or bad - which is much more in line with the normative choices scientists are forced to make in climate predictions, life cycle analyses et cetera without undermining their value.

Latour uses (in his usual witty manner) the example of the 'climate gate scandal' to examine the modernist worldview that most, including scientists, believe in so firmly - and finds that it is inadequate when attempting integrate science into political decisions on policy. He further proposes that rather than speaking of 'matters of fact' we should consider the object of science to be 'matters of concern', which allows us to appreciate the value judgments that scientists must make rather than dismissing them as unscientific.

Note: I am by no means an expert on Latour; I came across his texts recently in the context of morality implicit in technology, and found them both entertaining and insightful. I'm curious what others in this topic think of his stance.

I have never even heard of Latour but think the problems come in when someone lets their own 'opinions' outweigh the facts that they are seeing. Ala the 'climate gate scandal', the people doing the calculating didn't 'like' some results so just left them out, that is NOT good Science. What would have been better is if they were not necessarily included on the charts but were included in the wording and noted as 'anomalies'! Not including them anywhere made them look incompetent at best and deceptive at worst, thereby negating any good that could have come out of their research and Science! I think the basic theme is too EXPLAIN what you found and then EXPLAIN why you chose this over that, and that the failure to do that can result in anything you do being called 'junk' and seen as worthless! Especially when someone else is doing similar research and does find the data you 'left out'! Everyone makes 'judgment calls' every day, all day, but for a Scientist to just ignore data that could be counter to their research is just plain silly and bordering on incompetent! MOST Scientists KNOW their subject inside and out and KNOW when data is relevant or not, but to not record data that doesn't fit your expected outcome, is wrong and can lead to 'climate gate'. The old saying comes to mind...document, document, document! If ones documents it then it can be explained, it is then up to other Scientists to know if that documentation is valid or not.

Ver Greeneyes
Ver Greeneyes
Joined: 26 Mar 09
Posts: 140
Credit: 9562235
RAC: 0

Bruno Latour Yeah, the

Bruno Latour

Yeah, the scientists didn't handle things too well, but neither did the media. The problem is that even though normatively weighing various sources of data is an essential part of science, it doesn't fit in our worldview. Science is meant to be about cold, hard facts, indisputable truths, or so we would like to believe (Kuhn's paradigms be damned). And so the scientists tried to defend their numbers as completely factual and objective when they were clearly not, and critics used that as an opportunity to undermine the value of the research.

But things aren't as simple nowadays - most truly objective facts are very narrow in scope and inconsequential. Putting those facts together to form a larger picture, choices have to be made. Do you weigh ecotoxicity more heavily than carbon emissions? Or the other way around? There are no objective justifications for these labels, but they are essential for making useful statements. That's why 'matters of fact' should be 'matters of concern', to turn back to Latour (he's certainly not the only one who thinks this way - he just has a nice way of putting it :) ).

It's important for both politicians and scientists to understand this, or the scientific method will only lose more and more credibility. Like it or not, we still need these normative compositions to be able to understand the situation. And consensus, flawed as it may be, is still the best way we have to tell if a composition is good or bad. Of course, that doesn't take away from the fact that there need to be people willing to think outside the box every now and then, as long as they know what they're talking about!

mikey
mikey
Joined: 22 Jan 05
Posts: 11888
Credit: 1828071059
RAC: 206164

RE: Bruno Latour Yeah, the

Quote:

Bruno Latour

Yeah, the scientists didn't handle things too well, but neither did the media. The problem is that even though normatively weighing various sources of data is an essential part of science, it doesn't fit in our worldview. Science is meant to be about cold, hard facts, indisputable truths, or so we would like to believe (Kuhn's paradigms be damned). And so the scientists tried to defend their numbers as completely factual and objective when they were clearly not, and critics used that as an opportunity to undermine the value of the research.

But things aren't as simple nowadays - most truly objective facts are very narrow in scope and inconsequential. Putting those facts together to form a larger picture, choices have to be made. Do you weigh ecotoxicity more heavily than carbon emissions? Or the other way around? There are no objective justifications for these labels, but they are essential for making useful statements. That's why 'matters of fact' should be 'matters of concern', to turn back to Latour (he's certainly not the only one who thinks this way - he just has a nice way of putting it :) ).

It's important for both politicians and scientists to understand this, or the scientific method will only lose more and more credibility. Like it or not, we still need these normative compositions to be able to understand the situation. And consensus, flawed as it may be, is still the best way we have to tell if a composition is good or bad. Of course, that doesn't take away from the fact that there need to be people willing to think outside the box every now and then, as long as they know what they're talking about!

Very true too many people think Scientists 'know everything' about their subject and then when it comes up that they don't we often get mad and disappointed and blame them for our own beliefs! As I said though explaining why you chose what you did is very important to the process, peoples opinions on any subject may differ but the facts do not. On matters of Cosmology or Climate Change I am NOT an expert, so I must rely on others who hopefully can filter the good from the junk and make an 'informed' decision using the 'matters of concern', as opposed to me not being able to see the forest because those trees keep getting in the way!

Rod
Rod
Joined: 3 Jan 06
Posts: 4396
Credit: 811266
RAC: 0

Its simple.. All scientists

Its simple..
All scientists have to do is determine how things are..

Leave Why things are.. to the policy makers and the mob..

There are some who can live without wild things and some who cannot. - Aldo Leopold

Ver Greeneyes
Ver Greeneyes
Joined: 26 Mar 09
Posts: 140
Credit: 9562235
RAC: 0

Things haven't been that

Things haven't been that simple in a long time, certainly not since science became a tool of the state in the 18th century or so. If scientific activists hadn't entered the social arena, climate change activism would probably still consist of some mother earth hippies taken seriously by no one. It's exactly because scientists are people, and they saw that no one in politics cared about, say, all the stuff we were pumping into our rivers, that they started getting involved in more than just taking measurements and building models exclusively for use within some elitist scientific community. Which is not to say the scientific community at large always had eye for such things - worrying about the effects of dumping chemicals is only about 40 years old, before which everyone assumed it was okay.

These days, a lot of the sciences are inextricably linked with social expectations. For the Big Science projects, this is even true of particle physics - they couldn't get away with building the LHC without at least pointing out the jobs it would create, the advances in technology its construction would bring et cetera. If science is in desperate need of mediators just to get funding (where it was largely left to its own devices throughout most of the 20th century), do you really think politicians can be expected to use the output of scientific findings on their own?

What's more worrying is that politicians increasingly don't listen to scientists at all, rather choosing to go with what's socially favorable or what their gut says. Scientists themselves bear some responsibility for this as they adamantly hold that scientific fact is unquestionable, when the reality is that most fields are rife with contradictory results. On the flipside, that reality has lead to some asserting that science can be of no help at all, which is equally wrong. Science needs to be aware of its own strengths and weaknesses and learn to defend its results in the face of uncertainty, rather than denying it exists.

Matt Giwer
Matt Giwer
Joined: 12 Dec 05
Posts: 144
Credit: 6891649
RAC: 0

RE: Bruno Latour Yeah, the

Quote:

Bruno Latour

Yeah, the scientists didn't handle things too well, but neither did the media.

Let me repeat ... the MEDIA, as a matter of "responsible" journalism, has decided to take an advocacy position with regard to global melting. That is a conspiracy hiding behind high falutin' doublespeak.

Quote:
The problem is that even though normatively weighing various sources of data is an essential part of science, it doesn't fit in our worldview. Science is meant to be about cold, hard facts, indisputable truths, or so we would like to believe (Kuhn's paradigms be damned). And so the scientists tried to defend their numbers as completely factual and objective when they were clearly not, and critics used that as an opportunity to undermine the value of the research.

It can be expressed much more simply. Science must be devoid of value judgements. The world may become warmer or colder but calling it a disaster (or as I often suggest in rebuttal an improvement) is a value judgement.

A rational person cannot have an emotional involvement in the outcome of research. If one does then there will be another Piltdown Man. Back in Galileo's time a geocentric universe had an emotional involvement even though masked in theology and bible interpretation.

It is the same today.

I am not interested in looking it up but just a few days ago I followed a link saying regions were becoming drier because of global melting. The article however said that some places were becoming drier while others wetter. There was a global map showing both is red and blue. By Mk. 1 Eyeball there were more wetter areas than drier AND they bordered each other. Not a trend at all because they bordered east to west meaning equatorial regions were by eyeball at least breaking even and likely wetter. But the media conspiracy to advocate the cause of warming gave the headline the caught my attention. Why? If my eyeball is correct equatorial regions are getting more rain.

Comment viewing options

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