04 Apr 16
Originally posted by humyI would think the tank problem is something that could actually be tested with computers. ie a large number of tests could be run to see how closely predictions match reality.
I also intend to mathematically prove all those textbooks (and web links ) are wrong about the tank problem and publish that proof for all to see and scrutinize. I also intend to show the correct equation for calculating it and prove it valid. The correct equation gives approximately similar but not exactly the same probabilities.
04 Apr 16
1 edit
Originally posted by twhitehead
Given that you clearly accept that probability may vary based on perspective and that perspective is really about what information a person has, it seems unreasonable to not simply refer to a given probability based on what information was used to generate it rather than instead tying it to and individual or group and then calling it 'true probability'. I ...[text shortened]... I also find your rejection of infinities because you can't understand them to be highly suspect.
I disagree. It is a 'tricky concept' because it is wrong; and I intend to prove it and publish that in my book.
Well then you will fail. The issue is a definitional one and not something you can disprove logically.
we shall see. the issue is not just about definition; it is about definition with or without contradiction.
04 Apr 16
5 edits
Originally posted by twhiteheadIf that is done, I would claim it would almost certainly find my statistical methods predict reality more closely than the use of the conventional ones.
I would think the tank problem is something that could actually be tested with computers. ie a large number of tests could be run to see how closely predictions match reality.
But, actually no such tested should be needed; I have the proof via contradiction that their methods are wrong and mine must be the right one so I would take the act of testing it that way as a disappointing sign that mathematical or deductive proof is not enough for them. It would be a bit like the act of testing Pi is correct by making a close-to-perfect-circle-as-you-can-make-it round solid disk and then measuring its circumference and then its diameter and then dividing the two numbers to see if it approximately equals Pi; if you got the mathematical proof of something and you have thoroughly checked it is flawless so you are sure it is valid, there should be no need for such a check via an empirical observation.
05 Apr 16
Originally posted by humyThis kind of contradicts what you claim to be setting out to do. You claim you are revolutionising a branch of mathematics that includes rigorous mathematical proofs. Now it would be fine if you are just bringing a new way of looking at things and inventing a new language to describe it etc, but you are claiming that the current way is actually wrong ie that either the proofs themselves are invalid or there are some axioms that are false.
But, actually no such tested should be needed; I have the proof via contradiction that their methods are wrong and mine must be the right one so I would take the act of testing it that way as a disappointing sign that mathematical or deductive proof is not enough for them. It would be a bit like the act of testing Pi is correct by making a close-to-perfect-circ ...[text shortened]... you are sure it is valid, there should be no need for such a check via an empirical observation.
05 Apr 16
Originally posted by humyWill you be publishing your proof on vixra.org?
If that is done, I would claim it would almost certainly find my statistical methods predict reality more closely than the use of the conventional ones.
But, actually no such tested should be needed; I have the proof via contradiction that their methods are wrong and mine must be the right one so I would take the act of testing it that way as a disappointing ...[text shortened]... you are sure it is valid, there should be no need for such a check via an empirical observation.
05 Apr 16
Originally posted by KazetNagorraI might do. Have not really thought about it although I would have thought publishing it in a book would be 'sufficient'.
Will you be publishing your proof on vixra.org?
I certainly would want it to be published as a physical book; preferably with a hard-back option.
I assume I should also have it published as an e-book but haven't researched how to go about arranging that and don't yet know the first thing about e-books.
05 Apr 16
Originally posted by humyIf you do it through a publisher then they will do the ebook for you.
I might do. Have not really thought about it although I would have thought publishing it in a book would be 'sufficient'.
I certainly would want it to be published as a physical book; preferably with a hard-back option.
I assume I should also have it published as an e-book but haven't researched how to go about arranging that and don't yet know the first thing about e-books.
I doubt that you will be able to get a publisher unless you pay for it all. Mathematics books really aren't good sellers unless they are text books.
I recommend you finish your work and get it recognized before trying to publish.
05 Apr 16
1 edit
Originally posted by twhitehead
This kind of contradicts what you claim to be setting out to do. You claim you are revolutionising a branch of mathematics that includes rigorous mathematical proofs. Now it would be fine if you are just bringing a new way of looking at things and inventing a new language to describe it etc, but you are claiming that the current way is actually wrong ie that either the proofs themselves are invalid or there are some axioms that are false.
This kind of contradicts what you claim to be setting out to do. You claim you are revolutionising a branch of mathematics that includes rigorous mathematical proofs. Now it would be fine if you are just bringing a new way of looking at things and inventing a new language to describe it etc,...
...WITH mathematic proofs, WITH mathematical proofs.
but you are claiming that the current way is actually wrong
correct. But the very source of the error is not to do with what it has but what it hasn't; there is something missing.
ie that either the proofs themselves are invalid
not the mathematical proofs; they are valid, no argument there. All mathematical inferences are correct.
or there are some axioms that are false.
No, not 'false'. None of them are false; no argument there.
BUT there is one missing! I have called this missing axiom the tie axiom. Without this axiom, you get contradictions, paradoxes, problems (such as the problem of induction) and total nonsense because you haven't got sufficient constraints on what defines a probability and the result is wild freedom what to call a probability allows you to call a piece of total nonsense a 'probability' when it isn't a probability and then, when you input that nonsense as a 'probability' into your inferences, you output contradictions, paradoxes and many problems; rubbish in - rubbish out.
05 Apr 16
Originally posted by humyI still don't get it. You are saying the mathematical proofs are not false but they do have contradictions?
No, not 'false'. None of them are false; no argument there.
BUT there is one missing! I have called this missing axiom the tie axiom. Without this axiom, you get contradictions, paradoxes, problems (such as the problem of induction) and total nonsense because you haven't got sufficient constraints on what defines a probability and the result is wild freedom wh ...[text shortened]... ur inferences, you output contradictions, paradoxes and many problems; rubbish in - rubbish out.
What a 'probability' is, is a matter of definition. There is no 'true meaning' of the word.
05 Apr 16
Originally posted by twhiteheadBut my book won't be just about a bit of maths, it will revolutionize most of philosophy and most of science ( excluding pure mathematics which is the only area it will leave unaffected despite defining many new mathematical constants and functions -its all strictly applied maths )
If you do it through a publisher then they will do the ebook for you.
I doubt that you will be able to get a publisher unless you pay for it all. Mathematics books really aren't good sellers unless they are text books.
I recommend you finish your work and get it recognized before trying to publish.
05 Apr 16
5 edits
Originally posted by twhiteheadNo, these mathematical proofs are perfectly self-consistent i.e. they have no contradiction. Nevertheless, the conventional wisdom of probability leads to contradictions because there is something missing in it.
I still don't get it. You are saying the mathematical proofs are not false but they do have contradictions?
.
What a 'probability' is, is a matter of definition.
Correct. What if I were to prove that what is currently defined as a 'probability' leads to a contradiction because there is an important axiom left out? Insert that currently missing axiom in, all those contradictions disappear and that is the ONLY way to solve the contradiction thus proving via contradiction we MUST add that axiom to what we mean by probability else we are not being completely logically coherent. Thus it isn't just purely 'a matter of definition'; it is also the important matter of making our definition logically coherent.
05 Apr 16
Originally posted by humyWe will probably just have to wait and see your final work before discussing it further unless you can give an example of another area of maths where taking out one axiom leads to contradictions in what remains. It doesn't make sense to me, but as I say, without being able to see an example its not clear how it can be discussed.
Correct. What if I were to prove that what is currently defined as a 'probability' leads to a contradiction because there is an important axiom left out? Insert that currently missing axiom in, all those contradictions disappear and that is the ONLY way to solve the contradiction thus proving via contradiction we MUST add that axiom to what we mean by probabili ...[text shortened]... ter of definition'; it is also the important matter of making our definition logically coherent.
How long do you estimate before you are done?
05 Apr 16
Originally posted by twhiteheadI am afraid I don't know because my best estimate keeps changing. But my current best intuitive guess is I would be finished right at the end of this year BUT that is assuming my current work on producing generic equation and generic software to speed up my research ~5 fold will soon come to a stoppable point and with total success. If that plan fails or takes a lot longer, we could be looking at up to an agonizing long ~4-years! But I am cautiously optimistic that it won't come to that.
How long do you estimate before you are done?
05 Apr 16
7 edits
Earlier I stated that saying that a probability is "infinitesimally small" when you cannot say exactly what numerical value it is means it is not a probability and implied we shouldn't say it is an "infinitesimally small" probability because that would imply it is a probability, which it isn't.
But changed my mind because it has just occurred to me that as long as you emphasize and make clear that that said probability is not really a probability i.e. not a true probability but a pseudo probability, I cannot think how it could do any harm to say it is "infinitesimally small" as this would be just an issue of purely academic semantics.
05 Apr 16
Originally posted by humyAnd I still maintain it is a definitional issue. If I define 'probability' to include what you call 'pseudo probability' then problem solved. Certainly mine would be the standard textbook definition and yours would not.
But changed my mind because it has just occurred to me that as long as you emphasize and make clear that that said probability is not really a probability i.e. not a true probability but a pseudo probability, I cannot think how it could do any harm to say it is "infinitesimally small" as this would be just an issue of purely academic semantics.