The Flat Earth Society

Flat Earth Discussion Boards => Flat Earth Theory => Topic started by: edby on January 09, 2019, 12:35:30 PM

Title: Models vs patterns
Post by: edby on January 09, 2019, 12:35:30 PM
Some Flatearthers like to say that astronomical predictions are merely based on patterns, and that such patterns would exist whether or not the earth were flat. Hence RE = 0, FE = 0. It’s a draw.

But science itself does not measure success in that way. First, science requires an explanatory model. And it rates models in the following way. (1) Prefer a more accurate model to a less accurate one. (2) Given models of roughly equal accuracy, prefer the simpler model to the more complex one. (3) Prefer any model to no model at all.

So, take the pattern we observe of the sun rising, then setting, then a period of near darkness, then sun rising again. This is a well-observed pattern. Then the current scientific model is that this is caused by a relatively stationary sun, with the earth revolving on an axis or rotation. This is more than just noting the pattern, it is explaining it. An earlier model (the Ptolemaic) has a fixed earth with the planets, stars and sun revolving around it. Both models score roughly equally except for planetary motion. The Ptolemaic model does explain these, but it is more complex than the Newtonian system that replaced it, so prefer the current heliocentric one, which is simpler.

Then we have the Flat Earth model, which aims to explain the day-night pattern by a sun rotating around the flat disc of the Earth. How does this score? Well it’s a model of sorts. However, it does not accurately predict the angle that the sun appears each day, the position it rises in the east and so on. So it’s not really a model, and certainly not an accurate one. I have seen models that explain the observations in terms of light bending and so on, but (a) these are based on the RE system anyway and (b) they are far more complex.

In summary, it’s not about which system is ‘true’. Rather, which model best explains the observations. The current scientific model beats the flat earth one based on the scoring system proposed by Science.

Of course there may be other scoring systems. Perhaps someone would like to suggest one that favours flat earth models?
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 12:53:34 PM
But science itself does not measure success in that way. [...]
It's a fine allegation, and one that arguably favours you, so I'm not surprised you'd embrace it. Unfortunately, we've already established that this is often not the case. Just stating that it happens is worthless.

However, it does not accurately predict the angle that the sun appears each day, the position it rises in the east and so on.
These allegations are also warm and fuzzy, but they're better from the previous ones in that they're merely unsubstantiated, and not outright false.

Of course there may be other scoring systems. Perhaps someone would like to suggest one that favours flat earth models?
I propose that truth is a prerequisite, and that you're bending it so hard here it's about to snap. A "model" which is (more or less) consistent with itself but which is otherwise false is strictly worse than no model at all. This is in direct contradiction to your point (3). Deception is not superior to a lack of knowledge - and this is something that real science acknowledges.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 01:02:35 PM
But science itself does not measure success in that way. [...]
It's a fine allegation, and one that arguably favours you, so I'm not surprised you'd embrace it.
I am simply stating how established science measures 'success'. I ended my post asking for suggestions of scoring models in different ways. Over to you.

Quote
I propose that truth is a prerequisite
So your scoring system for a model is to prefer a true model to a false one? But how do we establish the truth? By observation presumably?

Quote
These allegations are also warm and fuzzy, but they're better from the previous ones in that they're merely unsubstantiated, and not outright false.
My allegation was that FE models do not accurately predict the angle that the sun appears each day, the position it rises in the east and so on. To test that statement, we need an FE model. Where is that model? If there is no model, then Science applies test (3) (prefer some model to no model). If there is a model, then its accuracy can be tested. Let's do that. I am happy to be proved wrong, Science is about that kind of being happy.

[EDIT]
Quote
A "model" which is (more or less) consistent with itself but which is otherwise false is strictly worse than no model at all.
No the scientific model is not consistent with itself, but rather with the observations it is trying to model.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 01:14:39 PM
I am simply stating how established science measures 'success'.
And I pointed out that your statement is false, and I explained why. "Simply stating" a falsehood is not helpful.

So your scoring system for a model is to prefer a true model to a false one?
No, this happens prior to scoring. There's no need to mark something that doesn't meet the submission criteria.

No the scientific model is not consistent with itself, but rather with the observations it is trying to model.
Another unsubstantiated allegation.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 01:17:11 PM
I am simply stating how established science measures 'success'.
And I pointed out that your statement is false, and I explained why. "Simply stating" a falsehood is not helpful.

So your scoring system for a model is to prefer a true model to a false one?
No, this happens prior to scoring. There's no need to mark something that doesn't meet the submission criteria.

No the scientific model is not consistent with itself, but rather with the observations it is trying to model.
Another unsubstantiated allegation.

Well as you are challenging one of the main views on scientific method, an area where I have published work, I suggest putting your challenge to one or more of my co-workers.

I can also provide plenty of citations to support my claims.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 01:18:51 PM
Well as you are challenging one of the main views on scientific method
Absolutely not. You are the one making the claim of superiority with unqualified and unjustified statements. We won't be having a "Prove me wrong" discussion here. Start backing up your claims.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 01:22:05 PM
Well as you are challenging one of the main views on scientific method
Absolutely not. You are the one making the claim of superiority with unqualified and unjustified statements. We won't be having a "Prove me wrong" discussion here. Start backing up your claims.
Read what I wrote. I am not asserting that the mainstream view on modelling is correct, merely stating what it is. You see the difference?
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 01:26:50 PM
Perhaps start with this paper. There are many.

https://faculty.wharton.upenn.edu/wp-content/uploads/2016/10/Simple-versus-complex-forcasting-The-evidence_1.pdf
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 01:29:53 PM
Read what I wrote. I am not asserting that the mainstream view on modelling is correct, merely stating what it is. You see the difference?
You are stating what you'd like it to be. Unfortunately, your allegation does not match the reality of the situation.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 01:32:50 PM
Read what I wrote. I am not asserting that the mainstream view on modelling is correct, merely stating what it is. You see the difference?
You are stating what you'd like it to be. Unfortunately, your allegation does not match the reality of the situation.
Take a look at that paper, and if that's not satisfactory, I can cite many more.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 01:35:16 PM
Take a look at that paper, and if that's not satisfactory, I can cite many more.
You continue to mistake stating something for proving it or establishing it as true. It's okay, we can wait.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 01:36:19 PM
No the scientific model is not consistent with itself, but rather with the observations it is trying to model.
Another unsubstantiated allegation.
I suggest you look at the statistical (and scientific) principle of 'expected vs actual'. 'Expected' is the result given by the model, 'actual' is the observation. The observation, if unbiased, is independent of whatever model we use. For example, the observation that the sun rises every day is an observation that is model-independent.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 01:38:15 PM
Take a look at that paper, and if that's not satisfactory, I can cite many more.
You continue to mistake stating something for proving it or establishing it as true. It's okay, we can wait.
I did not write that paper.

[EDIT]Here's another, not a paper but a clear statement of scientific method.

https://evolution.berkeley.edu/evolibrary/article/phylogenetics_08

Quote
The parsimony principle is basic to all science and tells us to choose the simplest scientific explanation that fits the evidence.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 01:39:09 PM
I did not write that paper.
And?
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 01:41:53 PM
Quote
The view that simplicity is a virtue in scientific theories and that, other things being equal, simpler theories should be preferred to more complex ones has been widely advocated in the history of science and philosophy, and it remains widely held by modern scientists and philosophers of science.
https://www.iep.utm.edu/simplici/
I am not claiming the method is correct, rather that it is a principle widely used in modern science.  The IEP corroborates my claim.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 01:42:43 PM
I did not write that paper.
And?
And I am stating something that is supported by that paper. See also the two other cites I provided. Citation requested, citation provided.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 01:48:11 PM
Citation requested
I didn't request a citation. I asked you to resolve the multiple contradictions in your statements.

I am not claiming the method is correct, rather that it is a principle widely used in modern science.  The IEP corroborates my claim.
And what does it matter if we already know that your claim is false? The popularity of your claim won't magically eliminate issues from it.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 01:49:10 PM
Also worth citing:
Quote
This view [i.e. on the preference for simplicity] has a long and illustrious history. Though it is now most commonly associated with the 14th century philosopher, William of Ockham (also spelt “Occam”), whose name is attached to the famous methodological maxim known as “Ockham’s razor”, which is often interpreted as enjoining us to prefer the simplest theory consistent with the available evidence, it can be traced at least as far back as Aristotle. In his Posterior Analytics, Aristotle argued that nothing in nature was done in vain and nothing was superfluous, so our theories of nature should be as simple as possible. Several centuries later, at the beginning of the modern scientific revolution, Galileo espoused a similar view, holding that, “[n]ature does not multiply things unnecessarily; that she makes use of the easiest and simplest means for producing her effects” (Galilei, 1962, p396). Similarly, at beginning of the third book of the Principia, Isaac Newton included the following principle among his “rules for the study of natural philosophy”:
Quote
No more causes of natural things should be admitted than are both true and sufficient to explain their phenomena.
 As the philosophers say: Nature does nothing in vain, and more causes are in vain when fewer will suffice. For Nature is simple and does not indulge in the luxury of superfluous causes. (Newton, 1999, p794 [emphasis in original]).

In the 20th century, Albert Einstein asserted that “our experience hitherto justifies us in believing that nature is the realisation of the simplest conceivable mathematical ideas” (Einstein, 1954, p274).
Again, I am not claiming that the principle is the right one. Rather, that it has considerable support in mainstream science.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 01:51:00 PM
Again, I am not claiming that the principle is the right one. Rather, that it has considerable support in mainstream science.
Ah! You're finally budging! We moved down from "this is how it works" to "this has considerable support". Perhaps soon you'll start addressing the actual issue.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 01:51:36 PM
And what does it matter if we already know that your claim is false? The popularity of your claim won't magically eliminate issues from it.
Again, and as stated above, my claim is not whether the scientific method is correct, but rather, 'this is the mainstream scientific view, widely held etc'. You are deeply confused.

I stated right there in the OP that alternative methods of scoring models could be discussed or proposed.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 01:52:45 PM
Again, and as stated above, my claim is not whether the scientific method is correct, but rather, 'this is the mainstream scientific view, widely held etc'. You are deeply confused.
I know what your claim is, and your repeated attempts at strawmanning me are not helping your case.

We have already established that your statements, made in the OP and quoted in my first response, are false. The remainder of your post collapses due to that. Please address this.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 01:52:56 PM
Ah! You're finally budging! We moved down from "this is how it works" to "this has considerable support". Perhaps soon you'll start addressing the actual issue.
No, read the OP. I said "Of course there may be other scoring systems. Perhaps someone would like to suggest one that favours flat earth models?"
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 01:54:17 PM
No, read the OP. I said "Of course there may be other scoring systems. Perhaps someone would like to suggest one that favours flat earth models?"
The fact that I didn't propose an alternative system does not preclude me from pointing out that your description of your own system is so critically flawed that it collapses unto itself. I can ask you to fix your claim without making a counter-proposal, which is what I'm doing.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 01:54:22 PM
We have already established that your statements, made in the OP and quoted in my first response, are false. The remainder of your post collapses due to that. Please address this.

OK let's change 'But science itself does not measure success in that way. First, science requires an explanatory model. And it rates models in the following way. (1) Prefer a more accurate model to a less accurate one. (2) Given models of roughly equal accuracy, prefer the simpler model to the more complex one. (3) Prefer any model to no model at all.'

"But mainstream science itself does not measure success in that way. First, mainstream science requires an explanatory model. And it rates models in the following way. (1) Prefer a more accurate model to a less accurate one. (2) Given models of roughly equal accuracy, prefer the simpler model to the more complex one. (3) Prefer any model to no model at all."

Does that help?

[edit] I added a second 'mainstream' to make it clear.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 01:55:28 PM
No, read the OP. I said "Of course there may be other scoring systems. Perhaps someone would like to suggest one that favours flat earth models?"
The fact that I didn't propose an alternative system does not preclude me from pointing out that your description of your own system is so critically flawed that it collapses unto itself. I can ask you to fix your claim without making a counter-proposal, which is what I'm doing.
Nope. Mainstream science prefers accuracy to lack of it, and if both models are accurate, then choose simplicity. That's perfectly correct.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 01:57:10 PM
Does that help?
It does not. (3) continues to be false, and the very example you provided in the OP (the use of astronomical patterns, which necessarily includes the Saros Cycle) contradicts your core premise. Adding a weak qualifier won't resolve this.

Nope. Mainstream science prefers accuracy to lack of it, and if both models are accurate, then choose simplicity. That's perfectly correct.
Mainstream science does not prefer an incorrect prediction to the lack of a prediction. I can't believe that I'm explaining this to an adult, let alone one that previously tried to falsely claim academic credentials.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 02:02:42 PM
Does that help?
It does not. (3) continues to be false, and the very example you provided in the OP (the use of astronomical patterns, which necessarily includes the Saros Cycle) contradicts your core premise. Adding a weak qualifier won't resolve this.
How does the example contradict my core premiss?

Quote
Nope. Mainstream science prefers accuracy to lack of it, and if both models are accurate, then choose simplicity. That's perfectly correct.
Mainstream science does not prefer an incorrect prediction to the lack of a prediction. I can't believe that I'm explaining this to an adult, let alone one that previously tried to falsely claim academic credentials.

If mainstream science finds the prediction incorrect, then it looks for a better model.
Title: Re: Models vs patterns
Post by: totallackey on January 09, 2019, 02:07:25 PM
We have already established that your statements, made in the OP and quoted in my first response, are false. The remainder of your post collapses due to that. Please address this.
First, science requires an explanatory model.
Errr, not hardly...

Plenty of instances of scientists just shitposting articles of speculation and enjoying the congratulatory attaboys from similar like-minded numbskulls...
"But mainstream science itself does not measure success in that way. First, mainstream science requires an explanatory model.
The second coat of paint does nothing to brighten the room...

Perhaps you should change colors instead, or perhaps even consult an interior decorator.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 02:08:55 PM
How does the example contradict my core premiss?
I refer you to the many discussions we've had before on this forum, many of which you participated in and are evidently familiar with.

If mainstream science finds the prediction incorrect, then it looks for a better model.
Irrelevant.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 02:26:59 PM
Newton’s formulation of the principle is also worth consideration.
Quote
Rule 1 No more causes of natural things should be admitted than are both true and sufficient to explain their phenomena.
As the philosophers say: Nature does nothing in vain, and more causes are in vain when fewer will suffice. For Nature is simple and does not indulge in the luxury of superfluous causes.

Rule 2 Therefore, the causes assigned to natural effects of the same kind must be, so far as possible, the same.

Rule 3 Those qualities of bodies that cannot be intended and remitted [i.e., qualities that cannot be increased and diminished] and that belong to all bodies on which experiments can be made should be taken as qualities of all bodies universally.
For the qualities of bodies can be known only through experiments; and therefore qualities that square with experiments universally are to be regarded as universal qualities… Certainly ideal fancies ought not to be fabricated recklessly against the evidence of experiments, nor should we depart from the analogy of nature, since nature is always simple and ever consonant with itself…

Rule 4 In experimental philosophy, propositions gathered from phenomena by induction should be considered either exactly or very nearly true notwithstanding any contrary hypotheses, until yet other phenomena make such propositions either more exact or liable to exceptions.
This rule should be followed so that arguments based on induction may not be nullified by hypotheses. (Newton, I. 1999. The Principia: Mathematical Principles of Natural Philosophy; A New Translation by I. Bernard Cohen and Anne Whitman. Berkeley: University of California Press, p794-796)

[EDIT] Rule 1 ("more causes are in vain when fewer will suffice") looks like a direct quote from William of Ockham. "Et pro istis est ratio ista quia 'frustra fit per plura quod potest fieri per pauciora'".
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 02:42:22 PM
Quote
Aristotle Posterior Analytics 86a32 That affirmative demonstration excels negative may be shown as follows. (1) We may assume the superiority ceteris paribus of the demonstration which derives from fewer postulates or hypotheses-in short from fewer premisses; for, given that all these are equally well known, where they are fewer knowledge will be more speedily acquired, and that is a desideratum.

Quote
Ptolemy (c. AD 90 – c. AD 168) ‘We consider it a good principle to explain the phenomena by the simplest hypothesis possible.’ Quoted in James Franklin, The Science of Conjecture: Evidence and Probability before Pascal. The Johns Hopkins University Press, 2001. Chap 9. p. 241.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 02:53:42 PM
Quote
Einstein: “our experience hitherto justifies us in believing that nature is the realisation of the simplest conceivable mathematical ideas” (Einstein, A. 1954. Ideas and Opinions. New York: Crown, p274).
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 03:06:14 PM
I fail to see how spamming quotes is going to progress your attempts at resolving the contradictions in the OP.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 03:06:59 PM
I fail to see how spamming quotes is going to progress your attempts at resolving the contradictions in the OP.
You asked for citations.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 03:07:57 PM
You asked for citations.
I did not, and I explicitly clarified that the last time you made this error.

Citation requested
I didn't request a citation. I asked you to resolve the multiple contradictions in your statements.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 03:11:47 PM
I didn't request a citation. I asked you to resolve the multiple contradictions in your statements.

One thing at a time. Your quote below refers to "unqualified and unjustified statements". I took this to mean my statements about the methods mainstream scientists (Newton etc) to choose between models.

If you didn't mean that, what did you mean?

You also asked me to '[back] up your claims'. I assumed you meant citation. If not, what did you mean?

Well as you are challenging one of the main views on scientific method
Absolutely not. You are the one making the claim of superiority with unqualified and unjustified statements. We won't be having a "Prove me wrong" discussion here. Start backing up your claims.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 03:14:04 PM
If you didn't mean that, what did you mean?
I refer you to my first response in this thread.

You also asked me to '[back] up your claims'. I assumed you meant citation.
A citation does not resolve a logical contradiction. Also, saying that others agree with you does not establish truth.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 03:18:09 PM
Also, saying that others agree with you does not establish truth.
You are so confused that there is not point in continuing this discussion. By all means consign to some other forum. We know the thread exists, and I have copied it.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 03:19:51 PM
You are so confused that there is not point in continuing this discussion.
No, you simply misunderstood my first comment, and instead of taking a step back to think about the fact that you've contradicted yourself, you double down on quoting Einstein.

It's a combination of a couple of simple mistakes and logical fallacies. Not too difficult to unravel. But one must first actually want to improve.
Title: Re: Models vs patterns
Post by: WellRoundedIndividual on January 09, 2019, 03:24:14 PM
Pete, can you clarify what the logical contradiction is? I believe you were the one to say it is a logical contradiction by interjecting your opinion that a false model is not superior to no model. And I believe it is a false equivalency to say that a false model is a deception.  A false model can be a deception, but a false model can also be based on lack of knowledge. Once that knowledge is gained, then the false model is either corrected or thrown out and replaced by a new one, which happens all the time in science. I am of the opinion that the word false in the OP is meant to mean an incorrect one, not meaning false as in a lie. AKA faulty, lack of data, or bad data, etc.

Can you also clarify what you mean by deception? Are you indicating nefarious groups? Or are you talking about cherry picking data or fudging numbers, or what?
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 03:40:01 PM
At last, a reasonable response!

First, allow me to correct the specifics of my point on "false model vs. no knowledge". edby's suggestion was that science prefers any model over no model at all. I believe this to be incorrect, because any model would include models that are already known to be wrong. It is superior to say we don't know how something works if the alternative is to propose that it works through fairy magic.

The point on deception is an important one. It is well-known that science is plagued with people trying to advance their own career rather than human knowledge. Some over-reactive journalists will tell you that "most" scientific papers are actually not reproducible, although I'm not sure I'd put that much stock in these reports. Nonetheless, I would say that deliberately falsified research is inferior to no research at all, once again blatantly contradicting edby's claim that any model is better than no model.
Title: Re: Models vs patterns
Post by: ChrisTP on January 09, 2019, 03:51:56 PM
I happen to agree with pete on this point I think, some published papers can be falsified and people will do and have done that in order to further their career or get a minute of fame for their findings. However, this is what peer reviewing is for and a lot of scientists are caught out with incorrect findings or data, be it on purpose or not. The point in peer reviewing is to correct any mistakes or false data that someone may have in their findings. As such it is an imperfect system, but it is a system regardless (at which point, I think this is better than nothing). Saying "magic" created something where you do not understand otherwise is obviously inferior to saying you just don't know. With that said you need a model in the first place to be able to realise it is incorrect and thus correct it.

With regards to scientists falsifying data to further their own agenda; with every person that does that there is also a lot of other scientists that will read through and claim bullshit just to get that step up as well. Afterall, who wouldn't want to correct a popular finding? It's pretty messy anyway.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 03:56:11 PM
edby's suggestion was that science prefers any model over no model at all. I believe this to be incorrect, because any model would include models that are already known to be wrong.
But if the model is known to be wrong, rule 1 kicks in. Prefer more accurate to less accurate. Thales (?) proposed that the world was made of water, trying to show how his proposal had explanatory power. Science at that time (7 century BCE?) preferred that to no model at all. Then better models came along, i.e. better in the sense that they were more accurate, simpler etc.

You have to apply the rules in order. Rule 1: go for the most accurate. Rule 2, if models are equally accurate, choose the simpler. Rule 3, prefer any model given by rules 1 and 2, to no model at all.
Title: Re: Models vs patterns
Post by: WellRoundedIndividual on January 09, 2019, 04:00:42 PM
Word choice affects how your point is perceived.

But did the science at the time know that  Thales proposal was absolutely wrong or did they just not have the knowledge.

The key point here is that your OP should state that a model based on limited knowledge is preferred over no model, and that models based on deliberate lies should be ignored outright.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 04:01:31 PM
I would say that deliberately falsified research is inferior to no research at all, once again blatantly contradicting edby's claim that any model is better than no model.
If the data is falsified, then the model is not accurate. It may explain the falsified data, but won't explain data accurate data.

On the meaning of 'accurate data', well there's a whole science of that. Happy to expand.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 04:05:22 PM
But did the science at the time know that  Thales proposal was absolutely wrong or did they just not have the knowledge.
The key point here is that your OP should state that a model based on limited knowledge is preferred over no model, and that models based on deliberate lies should be ignored outright.
I prefer to eliminate subjective judgments about 'knowledge'. Who knows whether we have knowledge or not.

All we need is a model and a (large-ish) set of accurate data, i.e. observations. The apply a test to see how accurately our model predicts the observations.

I gave an example in another thread on lighthouses. Make accurate measurements of the height of each lighthouse. Make accurate observations of the maximum distance the light can be see from a specified height. Then apply some mathematical model to try and explain these observations.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 07:18:17 PM
But if the model is known to be wrong, rule 1 kicks in. Prefer more accurate to less accurate.
What is the accuracy of something that doesn't exist?

If the data is falsified, then the model is not accurate. It may explain the falsified data, but won't explain data accurate data.
That doesn't make it preferable to no model. Please focus on addressing what's actually being said. If all you're going to do is strawman me, is there much point to having a conversation? You could be having it with yourself just as well.
Title: Re: Models vs patterns
Post by: edby on January 09, 2019, 07:41:55 PM
But if the model is known to be wrong, rule 1 kicks in. Prefer more accurate to less accurate.
What is the accuracy of something that doesn't exist?
Zero.
Quote
If the data is falsified, then the model is not accurate. It may explain the falsified data, but won't explain data accurate data.
That doesn't make it preferable to no model.
My point was, if the data is falsified, then the model is a false model, indeed not a model at all. So it is no model. To count as being a model at all, it must have some (perhaps statistically measured) success at explaining observations.

I’m a long way from understanding your underlying point, Pete. Clearly you don’t like the idea that Science prefers some model (a genuine model) to nothing at all. But that’s a fact. With no model, you merely have a claim. Is your worry that FE consists mostly of claims, with no underlying explanatory model? But that’s a fact. Take ‘bendy light’. FE has no underlying predictive model with mathematical functions to explain light bending in different circumstances. Unlike Science, which can explain pretty precisely, e.g. under what conditions light is refracted. I assume you did basic school physics?

Or take gravitational acceleration. Science both measures this well, and explains it well. FE has no theory which explains the observed differences at different latitudes and heights. (You will immediately bring in ‘celestial gravitation’. Very well: show me the mathematical model that underlies it).

Everything in Science is explanation, mere claims amount to nothing. That idea goes right back to Thales.

[EDIT]
I would say that deliberately falsified research is inferior to no research at all, once again blatantly contradicting edby's claim that any model is better than no model.

Falsified research is not research, for the same reason that a fake diamond is not a diamond.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 09, 2019, 07:59:09 PM
What is the accuracy of something that doesn't exist?
Zero.
Okay, then why would you prefer a model with zero accuracy over no model? Sounds like they're equal by rule 1.
Title: Re: Models vs patterns
Post by: edby on January 19, 2019, 12:42:27 PM
The disagreement is clearly over whether we should prefer some model to no model. Here are three threads where three different models are discussed, and where there is no FE predictive model. In each case, science prefers the predictive model.

Rowbotham Lighthouse experiment (https://forum.tfes.org/index.php?topic=11728.0) This shows that the simple model d = sqrt(7h)/2 predicts with reasonable accuracy the distance a lighthouse will be visible across water. I.e. reasonably accurate in the sense that there is a high correlation between prediction and observation. Pete made a series of objections on the grounds that correlation does not imply causation. Well perhaps, but Science is simply trying to make accurate predictions here, where ‘accuracy’ is defined in terms of statistical correlation. Er, how else would we define accuracy?

FE has no model whatsover to predict these observations. FE makes various claims about refraction, perspective or whatsoever, but claims are not predictions. There is no FE model that I know of which uses perspective to predict, with high statistical accuracy, the distance at which we can see a lighthouse.

Flight times and distance (https://forum.tfes.org/index.php?topic=10191.0) shows that the simple Haversine model roughly predicts how long a flight will take, given the latitude and longitude of the airport. No great accuracy is to be expected, given time on the runway, delays to schedule etc. But there is a strong correlation, hence a strong prediction, nevertheless.

The usual FE maps, by contrast, show massive errors in calculation. There are two FE objections (1) jet streams (2) that there is no FE map. In reply, (1) there is no FE jet stream model, so Science rejects the jet stream hypothesis until it sees an accurate model, and (2) the admission of no map is the admission of no model, QED.

None of this means that we must rigorously reject any FE claim or hypothesis. Merely, science fails to consider such claims or hypotheses until it sees a model that is more accurate, or at least as accurate but simpler. Nothing so far.

Free Air Correction (https://forum.tfes.org/index.php?topic=11909.msg180005#msg180005) This shows how the simple Newtonian gravitational model predicts observed free fall acceleration at different heights. Here, the measuring instruments are unbelievably accurate, and the statistical correlation is very high. So Science prefers that model.

FE hypothesises that the effect is caused by ‘Universal Acceleration’. Fine, nice to have a competing hypothesis, however there is no model. Moreover, and much worse, any such model is bound to be more complex than the Newtonian one, because it posits two effects on acceleration, rather than one, namely UA and something like standard gravitational acceleration. Yet the UA model exists precisely to avoid any standard gravitational model. Law of Parsimony requires that we reject any such model.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 19, 2019, 01:34:51 PM
Here are three threads where three different models are discussed, and where there is no FE predictive model. In each case, science prefers the predictive model.
Once again, you attempt to shift the subject. Please get back to me when you're ready to defend or amend your thesis that a known false model (of value 0, per your own claim) is preferable to no model (of value 0, per your own claim).
Title: Re: Models vs patterns
Post by: edby on January 19, 2019, 02:03:43 PM
Here are three threads where three different models are discussed, and where there is no FE predictive model. In each case, science prefers the predictive model.
Once again, you attempt to shift the subject. Please get back to me when you're ready to defend or amend your thesis that a known false model (of value 0, per your own claim) is preferable to no model (of value 0, per your own claim).
I believe I said above that a 'known false model' is not a model at all.  I used the phrase 'false diamond' to illustrate this, no?

I believe you are also 'shifting the subject', I think you mean ground shifting. Not at all. My central claim is that Science prefers some model to no model (including a 'false model'), and I gave 3 examples clearly demonstrating why Science prefers a model.

[EDIT] Your unwillingness to discuss cases such as these are noted. Clearly you are itching, nay bursting, to consign this thread to the nether regions. But as I said before, I have a record of its existence. The whole thread reveals much about the FE mentality.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 19, 2019, 02:16:58 PM
I believe I said above that a 'known false model' is not a model at all.
And yet you defend known false models in your diatribes. This is a contradiction that you'll have to resolve.

I believe you are also 'shifting the subject', I think you mean ground shifting.
I once again refer you to my previous post on meaningless semantics. Instead of mocking my use of English a second language, try addressing the issues that you've created for yourself.

My central claim is that Science prefers some model to no model (including a 'false model')
Thank you for retracting one of your faulty points, even if you refuse to acknowledge that that's what you're doing. Only a handful to go.

I gave 3 examples clearly demonstrating why Science prefers a model.
And at least one of them is a model we know to be false - so now you've introduced yet another contradiction into the conversation. Your previous examples (which I note you have now quietly abandoned) focused on debates which were either unresolved or lost by your side. Your argument is flimsy because it relies on you listing examples and claiming victory, even if your own examples speak against you.

But as I said before, I have a record of its existence. The whole thread reveals much about the FE mentality.
I don't see why you'd think I care. The very reason we move threads that break the debating standards to the lower fora is so that a record of them remains easily accessible. And the fact that you keep asking me to move the thread after I told you I have no intention of doing so is revealing, too ;)

This is almost as good as when you've claimed that an academia.edu profile makes you an academic.
Title: Re: Models vs patterns
Post by: edby on January 19, 2019, 02:22:12 PM
Please stop derailing this thread. To get back on topic.
I believe I said above that a 'known false model' is not a model at all.
And yet you defend known false models in your diatribes. This is a contradiction that you'll have to resolve.
Which 'false model' am I defending?
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 19, 2019, 02:25:49 PM
Please stop derailing this thread.
Demanding that you back up (or retract, which you thankfully started doing) your ridiculous assertions is not derailment - if you didn't want to discuss them, you had the option of not creating a discussion thread for them. I'm sorry you dislike accountability, but you're just going to have to deal with it. Your backlog is growing.

Which false model am I defending?
Classical mechanics and General Relativity, to name a couple.
Title: Re: Models vs patterns
Post by: edby on January 19, 2019, 02:25:54 PM
This is almost as good as when you've claimed that an academia.edu profile makes you an academic.
Can you also stop derailing the thread with ad hominem remarks such as these. I believe I said that academia.edu profile is a sort of Facebook for academics. I have hundreds of academic friends and colleagues who use the site. Thousands of other academics use it. It is merely a place for keeping working papers and records of publications, that's all.

My academic reputation and status depends entirely on my peer-reviewed work. End of.
Title: Re: Models vs patterns
Post by: edby on January 19, 2019, 02:27:21 PM
Which false model am I defending?
Classical mechanics and General Relativity, to name a couple.
Oh right. These models have great predictive power. That is what makes them a model, as I have asserted many times. Your claim that they are 'false' amounts to the claim that they have little or no predictive power. Please demonstrate this.

[EDIT]I don't believe I have ever mentioned General Relativity by the way. I have defended the predictive power of Classical mechanics a few times, see e.g. my post on Free Air Correction. Which part of this are you challenging?
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 19, 2019, 02:28:08 PM
Can you also stop derailing the thread with ad hominem remarks such as these.
Describing your past actions is not an ad hominem.

Your claim that they are 'false' amounts to the claim that they have little or no predictive power
No, it does not. We return to my first reply to this thread.
Title: Re: Models vs patterns
Post by: edby on January 19, 2019, 02:34:25 PM
Your claim that they are 'false' amounts to the claim that they have little or no predictive power
No, it does not. We return to my first reply to this thread.
Oh right, where you made the bizarre claim that in e.g. choosing between models A, B and C, our selection criterion should be whether the model is ‘true’ or not. That is so absurd it deserves hardly any comment. (A criterion is a form of observable and verifiable benchmark. Unfortunately theories do not come with the label ‘true’ or ‘false’ attached to them).
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 19, 2019, 02:38:06 PM
That is so absurd it deserves hardly any comment.
Very mature. Let's see if we can get you to realise your confusion.

Of course, theories do not come with a truth value allocated to them - but if we know that a model fits some observations but does not fit others, then we know that what's modelled cannot be the case in reality. This is where we disagree. I propose that a model which we know is broken should not be considered more valuable than the absence of a model.
Title: Re: Models vs patterns
Post by: edby on January 19, 2019, 02:43:35 PM
I propose that a model which we know is broken should not be considered more valuable than the absence of a model.
But we don't know that, e.g. the free air correction model is 'broken'. How is it broken? It is limited only by the precision of the measuring instruments, amazing as that is.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 19, 2019, 02:45:29 PM
But we don't know that, e.g. the free air correction model is 'broken'. How is it broken? It is limited only by the precision of the measuring instruments, amazing as that is.
Ah, but this is where you once again ignore the broader theory and focus on the one observation you like, pretending no other observations should be considered. Your gravitational model has to be considered as a whole.
Title: Re: Models vs patterns
Post by: edby on January 19, 2019, 02:51:33 PM
But we don't know that, e.g. the free air correction model is 'broken'. How is it broken? It is limited only by the precision of the measuring instruments, amazing as that is.
Ah, but this is where you once again ignore the broader theory and focus on the one observation you like, pretending no other observations should be considered. Your gravitational model has to be considered as a whole.
Millions of observations. In any case, Science still prefers the Free Air Correction model to, e.g. Celestial Gravitation, which is no model at all, moreover introduces assumptions (of the existence of gravitational attraction) which the UA thesis was meant to dispense with. Show Science a model for Celestial Gravitation that competes with Free Air correction, and Science will prefer it.

Note I say 'Science will prefer it' . I haven't said whether I prefer it. I am merely stating how Science goes about its business.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 19, 2019, 02:54:22 PM
Millions of observations.
I once again refer you to my point on meaningless semantics. Correcting tiny details in my phrasing does not advance your cause, it merely documents your desperate attempts at deflection.

In any case, Science still prefers the Free Air Correction model to, e.g. Celestial Gravitation
And, in a shocking turn of events, I once again have to bring up a past point you've failed to address - that me outlining your failures does not require me to bring an alternative to the table. Your tu quoque does not apply here.
Title: Re: Models vs patterns
Post by: edby on January 19, 2019, 02:58:20 PM
Millions of observations.
I once again refer you to my point on meaningless semantics. Correcting tiny details in my phrasing does not advance your cause, it merely documents your desperate attempts at deflection.
You: 'one observation'.
Me: 'millions of observations'
You: 'tiny details in my phrasing '

This is laughable, and enough is enough. I shall resist any temptation to return to this thread.
Title: Re: Models vs patterns
Post by: Pete Svarrior on January 19, 2019, 03:02:26 PM
You: 'one observation'.
Me: 'millions of observations'
You: 'tiny details in my phrasing '
Yes. I obviously meant one category of observation, which you chose to misinterpret. I'll be slightly more generous than yourself and assume that you simply didn't get my meaning - but then you had the simple option of asking.

The fact that plenty of observations contradict your model is what you should be addressing, but your best attempt at slithering away from this is attacking someone's use of language.

This is laughable
I agree - your dishonesty, inability to remain internally consistent, and finally poor attempts at taunting your opponent will join your attempt at faking your status as an academic when we next need to recall your behavioural patterns. We will laugh and laugh. And, as you astutely pointed out, it's all recorded and documented here.