User talk:Lakinekaki/Bios theory

From Wikipedia, the free encyclopedia

.........this article should probably be under Bios with disambiguation links for BIOS (in computers) and Bios (this page) and any other uses of term 'bios' that may appear (ie. in biology, etc.) Also, figures need to be added that show the difference between bios and chaos, but I could not do this in my first submission....this whole note to be deleted.... comment by Lakinekaki, moved to the Talk page by bikeable (talk) 21:17, 10 January 2006 (UTC)

Contents 1 Circle map !! ?? 2 Long discussion 3 NPOV 4 Controversial ? 5 Novelty 6 #REDIRECT Chaos theory 7 Factual dispute / confusing 8 Novelty/isometry 9 LInk to chaos theory 10 Quick comment 11 Suspect claims in the first paragraph 12 Apparent conflict of interest by the author of this article

[edit] Circle map !! ??

This article is throughly confusing. The example given A_{n + 1} = A_n + gsin(A_n) is known as the circle map, and was studied by Kolmogorov and Yakov Sinai V. I. Arnold in the 1960's. A description of the "Sinai's tongues" (the phase-locking regions of the map) shows up somewhere in Kolomogorov's multi-volume book on statistical mechanics (the term "chaos theory" hadn't been invented yet). (I have hi-res pictures on my art gallery, unfortunately with no explanation of what they are. There was also more work done on this eqn in France and the US in the mid 1980's; I went to a lecture given by P. Cvitanovic on the subject. Its rather weird to find it here, and to see its motion called "biotic motion". Maybe I don't understand something here ...linas 14:47, 13 January 2006 (UTC)

Please see also the discussion on Talk:Chaos theory where I point out that the circle map is "well-known" to describe the chaotic motion of phase locked loops, as well as mechanical systems comprised of a coupled rotor and a motor, and so it should be as no surprise that the heart, which resembles a PLL, should undergo chaotic motion that resembles a PLL. Please review the equations for the PLL in that article, and you will see that the circle map is a strong simplification of them. linas 15:49, 13 January 2006 (UTC)

Here we relate the process equation to the circle map in the case where W=0 . The phenomena that we have been discussing in relation to the process equation are not available or are hidden in the circle map. This is the quote from the first external reference. Linas, pleas tell me which part do you not understand, and I will try my best to explain it?Lakinekaki 03:54, 15 January 2006 (UTC)

We should continue discussion at Talk:Chaos theory, and not here. linas 18:13, 15 January 2006 (UTC)

[edit] Long discussion

There is a long discussion about Bios Theory and biotic motion in the Chaos theory talk page. It is contentious.

Readers of this page should note that as of 2006 the Mathematical Reviews does not mention Bios theory or biotic motion and that the only articles in the Web of Science that mention the topic are those written by Sabelli or his co-authors:

• H. Sabelli and L. Kauffman (1999). "The process equation: formulating and testing the process theory of

systems". Cybernetics and Systems 30: 261–294. 

• L. Kauffman and H. Sabelli (2002). "Mathematical bios". Kybernetes 31: 1418–1428. 
• L. H. Kauffman and H. C. Sabelli (1998). "The process equation". Cybernetics and Systems 29: 345–362. 
• H. Sabelli (2000). "Complement plots: analyzing opposites reveals mandala-like patterns

in human heart beats". International Journal of General Systems 29: 799–830. 

• A. Sugerman and H. Sabelli (2003). "Novelty, diversification and nonrandom complexity define creative

processes". Kybernetes 32: 829–836. 

• M. Patel and H. Sabelli (2003). "Autocorrelation and frequency analysis differentiate cardiac and

economic bios from 1/f noise". Kybernetes 32: 692–702. 

• H. Sabelli (1998). "The union of opposites: From Taoism to process theory". Systems Research and Behavioral Science 15: 429–441.  (this may be a stretch)

• S. Levy-Carciente and H. Sabelli and K. Jaffe (2004). "Complex patterns in the oil market". Interciencia 29: 320. 

XaosBits 22:40, 17 January 2006 (UTC)

• The information above is incorrect. Look at the Chaos Theory discussion page for more info.65.42.82.144 03:50, 26 January 2006 (UTC)

Sorry, I meant Science Citation Index and not Web of Science. XaosBits 04:40, 26 January 2006 (UTC)

In my view, XaosBits has provided ample justification for an AfD nomination on the grounds of non-notability. I have raised below serious questions connected to a possible conflict of interest on the part of the author of this article, and also a serious question regarding a hidden agenda on the part of the organization which apparently employs said author.---CH 07:58, 12 May 2006 (UTC)

[edit] NPOV

Considering that this is a contraversial subject, should not the presentation of Bios be done in a manner befitting the uncertainty of the validity of any of the claims of bios? I think it is great to discuss such things on wikipedia, but I am afraid a lay reader may incorrectly think that Bios is a scientifically accepted concept. I believe the opening paragraph should establish Bios's place in modern science. - JustinWick 16:23, 27 January 2006 (UTC)

Yep. linas 18:09, 27 January 2006 (UTC)

Ditto. It isn't clear what the standing is.--142.176.130.187 16:57, 31 January 2006 (UTC)

Yes, well, we had a long drawn out argument on this (see above, and in particular the chaos theory talk page). I personally do not have the energy to continue in further arguments, but if you edit this article to make it clear that Bios is far more hypothetical and imagined than it is real, I will support such edits. linas 04:11, 1 February 2006 (UTC)

Far more imagined than real. What does this mean? Lakinekaki 20:01, 1 February 2006 (UTC)

[edit] Controversial ?

Tetracube said that this theory is contraversial, so I looked in the dictionary and found an explanation:

Controversial - A dispute, especially a public one, between sides holding opposing views.

Could anyone give a reference of someone disputing bios theory?Lakinekaki 05:16, 2 February 2006 (UTC)

How soon we forget? I vaguely remember a long drawn-out knock-down drag-out fight over at Talk:Chaos theory, ohh, only a week or two ago. Perhaps you remember something different? Please, do not start fighting again. linas 00:46, 3 February 2006 (UTC)

Because you dispute bios theory, doesn't mean that scientific comunity disputes it. If you know someone (who published related work), give reference.

If you happen to dispute, say, evolutionary psychology, shell we put that in the Wikipedia article? NO! It is there because there were studies that dispute it.

What seems to bother you is that bios theory is not well known, and not many scientists do research in that field. This may be due to: a) they don't know about it, b) they do know about it but don't find it appealing, c) they do know it, but don't care about it, d) they know and care about it, but do not know how to use it, e) something else.

If what bothers you is that some biotic series are regarded by many scientists as chaotic diffusion, then we can make a note on the Chaos theory page that domain of chaos theory is being disputed by some scientists, because I may be able to find references for that. Is this the controversy you were refering to?

I think that much more accurate would be saying that theory is not well known or something of that kind. I got to think about the wording. Lakinekaki 02:10, 3 February 2006 (UTC)

I really don't want to have this argument all over again. You are intentionally ignoring the previous conversations, or you are intentionally misunderstanding them. Furthermore, you are rehashing the same topics, and so this conversation is not making any forward progress. What bothers me is that (1) there is still no clear definition of what Bios is, despite a very long discussion about this, and (2) every example of Bios that is cited is in fact a "well known" example that can be found in books on chaos theory, and furthermore, the cited examples come packaged with mis-interpretations and inaccurate statements. Please stop picking fights, and please stop pushing your POV. Please direct your energies into something productive instead of trying to figure out how to win an argument that cannot be won. linas 05:00, 3 February 2006 (UTC)

I suggest that you add a section on controversy as you see it. Otherwise, it doesn't make sense having that word in the article. People will probably want to know why is it controversial. (ps. it seems that we are not on the same wavelength, because I really can't understand why are you saying some things.) Lakinekaki 07:59, 3 February 2006 (UTC)

We already had this conversation in Talk:Chaos theory. If bios theory is about dynamical systems, then it is about mathematics, and the standards of mathematics may be used to understand it. – XaosBits 12:46, 3 February 2006 (UTC)

???????????????????????????????????????????????????????????????????

I suggest that you write that nonsense in the contraversy section. It will be easy for me to present the opposite view. Lakinekaki 14:40, 3 February 2006 (UTC)

This article is very misleading. On first glance, it appears that bios theory is a mathematical theory that is accepted by the mathematical community. All key papers share the same co-author, meaning no-one else has bothered to write on the subject, perhaps because of the reputability of(or lack thereof) the journals in which these articles were published. If this is a mathematical theory, please provide AMS (American Mathematical Society) subject classification. Otherwise, file under pseudomathematics and link to "cranks."

Your third sentence contradict the second one. Also, why do you assume that other readers will have less critical reading abilities than you do? Lakinekaki

[edit] Novelty

Let {1,2,3,8} be a time series. According to the article, if we choose r = 1.1 then this time series has 2 isometries. This time series has 4! = 24 distinct permutations

1, 2, 8, 3

1, 3, 2, 8

1, 3, 8, 2

...

All of them have two isometries with r = 1.1. If is a time series with k isometries within r₀, then all permutations of the time series will have k isometries. The reason is that the definition counts the isometries for all pairs, regardless of order. So no shuffle of the time series will have less or more isometries, as the number of isometries stays constant.

According to the article, no time series has the property of novelty.  – XaosBits 16:15, 3 February 2006 (UTC)

According to what you just wrote, you have no understanding of novelty. Isometries are counted for vectors of dimensions greater than 1. So, we compare for example vectors of dimension 2:

12, 23 and 38

their magnitudes are square roots of: 1+4=5, 4+9=13, 9+64=73. None of them will be isometric for your given radius of 1.1. 12 and 23 will be isometric for the radius of 1.5.

of your permutations, most (maybe all) will have 1 isometry (I am lazy to calculate all of them) for radius 1.5, and therefore the original time series will have isometries neither greater nor lesser than shuffled copies (nor novelty). This will be supporting evidence that the series is neither biotic, nor chaotic, but it is random.

I hope that this simple example helped you in understanding one of the fundamental features of the biotic processes.

This example is to serve for educational purposes only, because the time series of 4 data points cannot be used for any meaningful analysis, with any of the non-linear dynamics' methods - in chaos theory or bios theory.

Lakinekaki 16:45, 3 February 2006 (UTC)

XaosBits has a point. With the current definition of "novelty" as given in the article, there are no time series that have "novelty"; there cannot be, since the permutation of a series does not change the set of differences between pairs. I can't begin to imagine how to change the definition of "novelty" to make it a meaningful definition. I suggest that Lakinekaki either fix the definition, or present an example of a series with "novelty". linas 18:39, 4 February 2006 (UTC)

For dimensional space of m > 1 Did you miss to read this in the definition? Lakinekaki

By the way, this question/request/accusation is serious. If this isn't fixed soon, I will label the article as pseudoscience, and if this results in a revert war, I'll start administrative proceedings against Lakinekaki. linas 18:46, 4 February 2006 (UTC)

How can you claim this in spite of the evidence to the contrary - look at the graphs below that demonstrate novelty as used in the definition. I suggest that we ask for arbitration because I will ask for it in the case that you start misusing your administrative privilleges. Lakinekaki

I can start objecting to the chaos theory definitions that are totally not testable or applicable on empirical time series and therefore render theory useless.Lakinekaki

Let me try and restate, but first a few definitions. Two vectors a and b are isometric within ε if their norms differ by less than ε:

.

Given a scalar time-series X with N elements { x₁, ... , x_N }, an m-dimensional single-step delay embedding of the time-series is a sequence U_m of m-dimensional vectors

constructed from the values of the time-series X. For the sequence of delay-embedded vectors the number of isometries C within ε can be counted using the Heaviside step function Θ, namely

.

The time-series X displays novelty with respect to a permutation σ of the elements of the time-series if the permuted time-series σX has a larger number of isometries within ε, that is,

.

Questions:

1. Is it C_σX > C_X or C_σX < C_X?
2. Is novelty a property of the time-series or of the map that produced the time-series?
3. Does a time-series display the property of novelty if it has novelty with respect to all permutations or just most permutations?
4. How is m chosen? A time-series may display novelty for one value of m, but not others. For example, if m= N-1 the number of isometries is fixed to zero or one regardless of the permutation.
5. Are there any limits in the definitions? How is ε chosen? If ε is too small there are no isometries and if ε is too large all pairs will be isometric.
6. Shouldn't the definition use a fraction of the pairs of vectors?

Novelty seems to be such a central concept in Bios Theory that I feel it should be clearly explained. I have added a tag to the article alerting readers of the difficulties.  – XaosBits 15:22, 4 February 2006 (UTC)

I see what you mean. These are good questions. I'll try to answer to all of them.

1. C_σX > C_X
2. novelty is a property of the time-series
3. this is in a way simmilar as with the second law of thermodynamics. It is probable that, for example, you get exactly the same order after shuffling (this case is not excluded by the shuffling operation!) that you had before and the series would be random by this test, but if the series is not really random this is highly improbable. I have been using Bios Data Analyzer for quite a while, and it is amazing how consistent are the results. In the simmilar way, there is probability (thou so small) that generated random series are chaotic or biotic or ordered, and shuffling of this series would then show results that would not be correct, but this is also highly improbable. (i.e. as it is highly improbable that there is some distant galaxy where you and I live on the same kind of Earth like planet, but you are defending this bios theory, while I am arguing against it :o) So, answer to you question would be in my oppinion, to most permutation.
4. To test the time series for novelty, it should have at least few hundred numbers. You can not really do this analyses with 30 numbers. You really can, but the results will not be very meaningful. If you have for example 500 numbers in the series, You can for sure test novelty for m=2..50, (increasing m increases the difference between non-random and shuffled series) and it will show consistent results. You can also test it for m=499, but when you get 0 or 1 isometries, you will probably want to experiment with other m's and radiuses. Common sense comes in play here!
5. Of course these apply.
6. I don't really understand this question. Can you reformulate it?

Lakinekaki 16:53, 4 February 2006 (UTC)

From answers 3, 4, and 5 it appears that an infinite series is needed. Thermodynamic quantities are only sharp in the infinite volume limit. And statistical concepts such as most permutations or majority of permutations may require large N to be well defined. If there is an infinite series then the count C of isometries within ε may not be well defined, as in most cases it will be infinite and therefore cannot be tested for novelty. Hence my question 6, about considering not C directly but the ratio of C and some other quantity.

Also, if novelty is a property of a time-series, how is the property of novelty associated to a system? I imagine that a system would have the property of novelty if its behavior can be seen as a dynamical system that produces a time-series that displays novelty. But which time-series? Is it for a typical initial condition? For some initial conditions?

It would be very helpful is someone familiar with the correct definition of novelty could refine my attempt.  – XaosBits 22:00, 4 February 2006 (UTC)

I don't quite understand why you think that infinite series are needed. I agree that large N may have to be well defined, and I'll look into that. You make however a good point about ratio of C and some other quantity. Bios Data Analyzer does calculate the ratio of C and the number of all possible isometries, as if all vectors were isometric with each other. I don't know if this is necessairy for definition of novelty, and I will have to check that. In my oppinion it should not make any difference, because we never deal with infinite series for obvious reasons.

As far as your question about whether the property of novelty is associated to a system - I don't know the answer. I think that there is a hypothesis that the system has to have bipolar feedback in order to generate novelty. I think that this question is very interesting, and that it's exploration may result in an interesting research.

Your question about initial conditions is confusing me. You can analyze logistic equation, and then process equation (you can do this analysis for various initial conditions in Excel), and then compare the results. Maybe after spending some time by analyzing this equations you will be able to answer your question, because I just don't understand it. I tend to think that logistic equation is always logistic equation, no mather what initial conditions you have. It's behaviour depends on its parameters. Same is true for process equation.

Lakinekaki 23:57, 4 February 2006 (UTC)

The infinities usually make the definitions cleaner, but they are not necessary. I am having difficulties in determining the novelty of a time-series. I started out with the map

f(x) = x + 4.62sin(x).

According to a figure in the Bios Theory article this should be a map that produces a time-series with the property of novelty. I choose the initial condition 0.4. This produces the time-series X with

• x₁ = 0.4
• x₂ = 2.1991...
• 
• x₅₀ = 8.8227...
• 
• x₅₀₀ = 20.4729...

The calculations where done with infinite precision and are accurate as displayed. If the calculation is done with machine precision the numbers will be a different time-series Y. For example, in my machine x₅₀ = 17.2866 (or better, y₅₀ = 17.2866).

For this time-series, C_X(ε = 0.01, m = 5, 500) = 126. On the machine precision time-series C_Y(0.01, 5, 500) = 85. If I try several random permutations σ_iX of the series X and compute C_σiX(0.01, 5, 500) for the different σ_i I get

149, 133, 143, 125, 141, 124, 117, 139, 141, 121, 140, 116

The count of isometries went up in seven cases and down in five. When I try the calculation with m = 3 most counts go down with a shuffle. When m = N - 1 the count is one, so the variation with m on the count is not monotonic. The results are also different for the numerical time-series Y.

If novelty is a property of time-series and the counts seem so sensitive to numerical approximations, it is not clear that the map discussed in the article displays novelty. As there is no clear definition of novelty, one cannot decide if X displays novelty or not. – XaosBits 16:05, 5 February 2006 (UTC)

I will put explanation and plots of embedding plots in next 2 days that display isometry quantification for dimensions 1..K whathever K you choose. It will make things much more clear. You will be able to see obvious differences between novelty and absence of novelty. Lakinekaki 18:01, 5 February 2006 (UTC)

A definition of novelty may be useful. As the examples given earlier show, it is not sufficient to say that a time-series displays novelty if the number of isometries increases under shuffling. It is not sufficient because m has not been specified nor has the radius ε. Picking different choices of m leads to different behaviors for the number of isometries, as in the earlier example. If for a fixed time-series, the fraction of isometry increasing shuffles, in say 100 shuffles, is plotted against m, then that function is not monotonic, making it unclear what value to choose for m.

No clear definition of novelty seems to exist. Because the section on novelty remains unclear and so difficult to verify, I am labeling it significantly inaccurate to alert readers. Maybe someone will provide a definition and remove the note.  – XaosBits 04:49, 6 February 2006 (UTC)

Can you make one thing clear for me:

• 1) do you claim that no clear definition of novelty exists - you don't think it is clear enough?
• 2) this article has no clear definition of novelty which may exist and I should find it?
• 3) you think that definition of novelty is not applicable to experimental time series that you presented?

Lakinekaki 05:55, 6 February 2006 (UTC)

1) Yes. 2) Yes. 3) Can't tell, because of 1+2. The current definition of "novelty" in the article is broken. According to the definition given, no time series will ever have "novelty", since permuting the elements of a set cannot change the distance between them. linas 06:39, 6 February 2006 (UTC)

I will demonstrate the opposite tomorrow. Now I will go to sleep. :0) Lakinekaki 06:45, 6 February 2006 (UTC)

The definition I have read is insufficient to ascertain if a discrete dynamical system, such as the process map, produces novelty. I gave examples of this earlier on.

The discussions of novelty I have read do not discuss the fluctuation of the number of isometries with different shuffles, an issue that could be resolved if the behavior of some isometry ratio is similar to that of thermodynamic quantities. That approach would require original research and an explanation of the order of the limits for the quantity C(ε, m, N). My impression is that unless a careful joint limit is taken the increase in isometries disappears.  –  XaosBits 12:44, 6 February 2006 (UTC)

I'm uploading some figures to make this novelty thing more clear. These are embedding plots that show the isometries for the series (bold lines) and shuffled copy (thin lines). The same shuffled copy is used in all figures. Bios Data Analyzer (BDA) randomizes sequence once, and uses that copy for all analyses. Each time you choose another series, or regenerate the same one, new randomization is done. The only difference between these plots is radius used, which increases by 10 each time. I put the last one showing the embedding plot of the sine wave just to illustrate that this method really distinguishes different patterns.

As you can see from these figures, we start with very low radius that does not give any useful information. This is one extreme. The other extreme is very high radius where both shuffled and original series show 100% isometry at each embedding. Does this mean that this method or definition of novelty is useless? Let's analyze further. Out of 5 plots, I actually succeeded of finding radius that will show novelty for all embeddings even though we were dealing with extremes here by analyzing all embeddings that in my oppinion is interesting only as a way of testing the limits of this method.

XaosBits would probably object now by saying that definition should give us a way to know this radius for any time series, of any length, and for any embedding. I would be very happy if he succeeds in improving bios theory by doing that. Until then, I will just have to be happy by acknowledging that bios is an experimental science, just like chaos. I will come to this later (you will love it :O).

Other objection would be that he can generate a permutation (i.e. by switching 7th and 43th data points), where this method would fail! Definition does not mention permutations. It mentions shuffling.

There is a third objection - how well novelty applies to the process map. Will discuss this further down.

Note that in the example given by XaosBits we are dealing with relatively short series and are testing isometries for extreme cases where number of embedding dimensions approaches the length of the series N. Having m = 497 doesn't really make any sense because you can only do 6 comparisons or so. Do you want to make any conclusions based on this? On the other hand, when m is closer to 1 you can do ~O(n*n) comparisons which gives much more statistically reliable results.

The purpose of developing BDA was to have a tool for analyzing very long series. User can change maximal m, radius, number of randomly compared vectors. Why this last is important you may ask. Because if you have N = 10000 data points, and you want to analyze dimensions m = 1..1000, it would take very long time to calculate all posibilities O(N*N*(maximal m)*(some other factors)). Instead, you can specify to calculate for example only 1% of these, and still get very reliable data that does not change much from calculating 100%! This also depends on the radius used. Experiments have shown that larger the radius, smaller this percent can be.

Almost all the questions XaosBits raised, we encountered while developing this software. As you can see from this plots (and the purpose of BDA is precisely this, to make things clear and to allow researcher to explore results by varing the parameters), novelty is very distinct from, for example, periodicity shown in the last figure.

There are 2 questions asked here: one is - have I provided you with good definition of novelty, and the answer is - yes, I really think so. The other is - is that definition applicable to the real world and empirical time series. Let me answer this one by entering into the area which is so dear to you. This is actually more fundamental question regarding the application of mathematical models to the real world phenomena. I will now explore how a hundred years old science, well established, with rigorous mathematical definitions that are clear and deal with infinities, science practiced by thousands of researchers, science called The theory of Chaos, how this science is approaching the same problem. Once we understand the answer to this question, we will be able to make some parallels between Chaos theory, and a very new science called Bios theory.

Lets try to apply the same rigor using chaos theory definitions to this very same series, and show if it is chaotic? (because after all, that is what you claim all this time for the process map biotic phase!)

First definition sais - sensitivity to initial conditions.

Lets start with the simple one (1,2,3,8). Is the series chaotic?

--Your answer here--(hint: Lyapunov exponent)

Now lets apply this to the numbers that you generated with the process map. Pretend for the moment that you don't know what is the map but have only data series. What is the minimal number of points that will give correct answer? Do you think that I could not find part of the series that may fail your tests?

--Your answer here--

Second definition sais - it must be topologically mixing. Can you tell me the value R, such that, for all n > R definition holds? Do you need to have a very long series to test this? What could be the minimal n?

--Your answer here--

Third definition sais - its periodic orbits must be dense. (A is dense in X if for any point x in X, any neighborhood of x contains at least a point from A.) Can you really apply this to the series you presented here? Do you need infinite series? If yes, how is this definition applicable and useful? If yes, how is it that so many empirical series are claimed to be chaotic?

--Your answer here--

I am sure that chaos definitions are very mathematically clear in your oppinion. Are they applicable?

There are many things I don't understand (and don't agree) about chaos theory and its application. Does this mean that I should put a factual dispute tag there? No, because I am sure that those definitions are useful in some applications and domains.

Finally, you seem to be objecting to the validity of the theory. That is OR.

Lakinekaki 23:51, 6 February 2006 (UTC)

I suggest that you stop trying to pick an argument, and stop trying to duck the issue. The above post does not define "novelty", which is the thing that is being debated. I will happily explain what topological mixing is on the talk page for topological mixing. I can explain the term "dense set" on the page for ergodic theory or dense sets or wandering sets, if you wish. However, the issue at hand is "novelty", and I was hoping that you would provide a definition for "novelty". Just to be clear: uploading a bunch of graphs and saying "here, this is novelty" is not a definition. linas 02:21, 7 February 2006 (UTC)

There is definition of novelty but you don't seem to like it. I can't help you with that.

If I wanted to pick an argument, I would repost my question about application of these definitions to the time series XaosBits presented here on the pages that you specified. However, I am not going to do this. I already wasted to much of my life time on these discussions. Lakinekaki

Where is the definition of "novelty"? As stated multiple times above, the definition currently in the article is non-sensical. By this, I mean that, according to the definition currently given in the article, no sequence can ever have novelty. Can you fix the definition so that it makes sense? linas 04:12, 7 February 2006 (UTC)

The answer to your questions to XaosBits is that the sequence is not chaotic, it has no Lyapunov exponent, its not mixing, and its not a dense set. linas 04:15, 7 February 2006 (UTC)

If you are refering to the process map sequence, than this means that most of the things regarded as chaos, and described in the chaos literature for decades are actually not chaotic, and therefore much of the science and discipline of chaos (specially its application) is nonsensical. Thank you for your answer. Lakinekaki

Some points in random order:

• A permutation is the standard term for the description of order of elements in a list with respect to a base order. A shuffle is the standard term for the re-arrangement of the order of the cards in a deck, and, by analogy, to other sets. For every deck shuffle there is a permutation that describes it. A permutation could be the swap of two of the numbers of the time-series or an interchange of all of them. There are n! permutations and there are n(n-1)/2 swaps.
• When trying to determine a parameter, such as the Lyapunov exponent, associated to a time-series, the assumption is that there is a dynamical system that produced the time-series. The measure associated with the dynamical system is seen as a probability measure and the data tested with traditional statistical methods (maximum likelihood, Bayesian techniques, ...). To carry out this procedure requires a definition of the parameter for the dynamical system so that definition can be used to create a statistics to apply to the time-series (an example of a statistics would be embed the time-series and estimate the Lyapunov for the reconstructed system). What is missing in this discussion is the definition of novelty for a dynamical system.
• From what user Lakinekaki wrote earlier, a reader may get the impression that determining novelty for a time-series consists of choosing one permutation and plotting around until the permuted time-series comes out with a higher count of isometries. That may be a procedure for a time-series but it is not an acceptable procedure for a dynamical system. The approach taken earlier by Lakinekaki was to generate a time-series for a map and then test the time-series. What if a much longer time-series had been taken, would the results be the same? What if a ballistic trajectory had been chosen to generate the time-series? What if the BDA program had picked a different permutation? What if the map had an even larger sensitivity to numerical approximations than the process map?
• There are several regimes for the isometry counts. In the small m regime the number of isometries goes down when the time-series is shuffled. So if we take N to infinity first no time-series will display novelty. I know all this would qualify as original research, but it is easy to check analytically for hyperbolic systems.
• It is easy to generate random sequences of numbers that have minimal number of isometries. The method of hunting around for ε and a permutation that increases the isometry count would indicate that this sequence displays novelty.
• The number of isometries within ε (or r) is a non-decreasing function of ε. I would like to read an explanation of why a user need not worry about a method that keeps increasing ε until something becomes large.
• I am assuming the vertical axis of the figures indicate C_X (bold line) and C_σX (thin line).
• In Mathematica I can count the isometries of a list with ten thousand elements for all the embedding dimensions from 1 to 1000 in about 1080 seconds, about 18 minutes of CPU time, using a laptop. Not that long.

 – XaosBits 22:06, 7 February 2006 (UTC)

Few notes. How can you know the parameters of the economical dynamical system that generates DJIA? If you cannot know this parameters, how can you (chaos literature) claim that it behaves chaotically if basic definitions of chaos cannot be tested on empirical series? Google search returns 32,800 results for 'chaos OR chaotic DJIA'. Is all that just nonsense or what?

I am glad to see that you have so much spare time in your life as to spend 18 minutes for that calculation plus 18 more for calculation of shuffled series (36 total). BDA gives very good results for these two in about 3 minutes (12 times less). I think that many researchers not familiar with Mathematica, and with less spare time than you, will find BDA quite usefull. Specially if they want to test for more than just one shuffle! By using BDA you can see that for many many many different shuffles (permutations if you like) you get (quantitatively) similar results (qualitatively the same). :O)

User doesn't need to keep increasing ε until something becomes large. After extensive testing, we found ε (as percent of the range of the series) that for most series analyzed doesn't fall into the extremes, but gives the user useful information right away! If user wants to test the limits, he can easily change ε. Maybe those extreme ε's can provide some useful information about the dynamicial system itself. Who knows. Future research will tell.

Lakinekaki

[edit] #REDIRECT Chaos theory

Given the various comments above, and that it read like gobbledegook to me, and that it seems to be non-notable: google gives on 264 hits [1] the top one of which is wiki, the second is User talk:Lakinekaki, the third is politics... and so on. Its just not notable. So rather than the tedium of VFD, redirect to the clostest thing seemed the best idea. Comments? William M. Connolley 19:29, 9 February 2006 (UTC).

I put some comments on your talk page. Can you cite your reference for notability policy of Wikipedia? I would like to read it again. Lakinekaki

Errrm, are you really not aware of Wikipedia:Notability? You should be. Also, I've pasted this, since it seems a shame to put it on my talk page where other people won't see it:

ERRRRRRRMMMMM, seems that you cannot read! I asked for policy not an essay!

This is an essay representing the opinion of some editors but by no means all or even most editors. This is not a policy or guideline. For a number of notability-related guidelines, see the infobox at the right. it sais further down ...that its relative obscurity does not make it unencyclopedic or preclude it in any other way....Lakinekaki

Wikilawyering won't help, nor will selective quotation: 'However, since Wikipedia is not a bureaucracy, there is not a strictly limited set of criteria for deletion. Articles are deleted on grounds of notability on a daily basis, and this has been common practice for over a year now.William M. Connolley 20:28, 9 February 2006 (UTC)

Oh, and have you read this? William M. Connolley 20:33, 9 February 2006 (UTC)

Yes I did, indeed I wrote it. That article is about software which is quite a different thing. And Software notability is at least proposed policy, not just an essay.Lakinekaki

Thank you for redirecting Bios Theory to Chaos Theory and by doing that disabling visitors to read extensive discussion on Bios Theory talk page. If you make few more blatant moves like that, I will report your case to the arbitration committee. Lakinekaki

So all in all, I've reverted back to redirect William M. Connolley 20:01, 9 February 2006 (UTC)

So all in all in all I'm reverting again, and submitting your actions to arbitration committee. Lakinekaki

In addition, if someone types 'Bios Theory' (obviously knowing about it) in wikipedia and ends up on 'Chaos Theory' page, that will be highly missleading and confusing for the reader.Lakinekaki

Please don't waste their time. They will laugh at you. William M. Connolley 20:28, 9 February 2006 (UTC).

I strongly oppose any redirect. Bios Theory, may not be mainstream, it may be small, but it HAS received government grants, it HAS had papers published in referred journals and at least one author (Kauffmann) is notable for other work. This the pattern you would expect for an emerging academic field. Notability criteria apply for Bands, Characters from Fiction, but for other work (from WP:N):

If an editor says that a "non-notable" article should remain, he or she may mean that its relative obscurity does not make it unencyclopedic or preclude it in any other way.

"Non-notable" is generally a non-NPOV designation.

The article in not vanity (not written by author), is not original research (it reports on original research), it is verifiable. It is factually correct (in that it is more or less accurately reports on the work), if third parties find factual errors in the theory then that should be reported in a criticism section, for us to critique the theory would be OR. --Salix alba (talk) 01:35, 10 February 2006 (UTC)

There was a long discussion in the Chaos theory talk page regarding the lack of citations of Bios theory. A few other points:

1. The article does not reflect the few articles in the literature (all peer-reviewed ones are co-authored by Sabelli). For example, the article's explanation of novelty is only partially the one given by Sabelli.
2. The author of the article also repeated some of the errors in the articles in the literature. For example, the transition point is incorrect in one of the figures.
3. The article gives the impression that Bios theory is a mathematical theory. The literature of the field fails to be mentioned in Mathematical Reviews, that is, none of the over 11000 reviewers working in any given year took up any of the articles in Bios theory. This may be a consideration when establishing a redirect. – XaosBits 04:04, 10 February 2006 (UTC)

I moved the comments of Lakinekaki (now below) from the middle of my point. Unlike Usenet articles, the guideline for the Wikipedia is to proceed descendingly. I felt the argument lost some of its relevance by being broken up. The paragraphs below were interlaced with the number paragraphs above, starting before paragraph 1. Normally discussions should not be tampered with and I am trying to undo the modifications made to my comments while preserving the new content. The original page is still available. – XaosBits 17:44, 10 February 2006 (UTC)

• Society for Chaos Theory in Psychology and Life Sciences is 6th result in Google out of about 25,000,000 for "chaos theory". I guess this means something. I am saying this because "Nonlinear Dynamics, Psychology, and Life Sciences" is published by that society, and you don't seem to have any good oppinion about that journal. It may not be indexed in ISI, and that may be the reason why you cannot find citations, but you cannot simply dismiss social sciences because Journal Citation Report produces Immediacy and Impact factors for the Science edition journals (4500 journals), but not for the Social Sciences Citation Index (1400 journals). Lakinekaki

1. I ask you to help me then by editing this definition as to reflect better Sabelli's articles.Lakinekaki 04:35, 10 February 2006 (UTC)
2. I have corrected this now.Lakinekaki 04:30, 10 February 2006 (UTC)
3. I acknowledge this and have renamed section to Mathematical bios. (Mathematical theory was there because I used Chaos Theory article as a template. It was not my intention to force any impressions.)Lakinekaki 04:19, 10 February 2006 (UTC)

I don't think Lakinekaki has a clue what this is all about. I've restored my bold redirect. William M. Connolley 20:58, 14 February 2006 (UTC).

[edit] Factual dispute / confusing

I removed factual dispute tag and left confusing one, because that seems to be reality, that some readers are confused. If you think I presented something false in the Novelty section, feel free to edit it and correct it. Lakinekaki 01:55, 10 February 2006 (UTC)

I restored the Factual Accuracy box. I have several reservations with the concept of novelty, as you may see from the above discussion, but that is not why I added the tag. The tag is there because it does not reflect what is in Sabelli's articles. –  XaosBits 03:14, 10 February 2006 (UTC)

? Why don't you put the definition that does reflect Sabelli's articles ? Lakinekaki 04:13, 10 February 2006 (UTC)

I added some info and rewrote little text in Novelty section. XaosBits, please tell me if you find any more objections here. Lakinekaki

[edit] Novelty/isometry

I restored the "factual accuracy" box, because this section remains vague about what its talking about. Let me be specific.

The article states:

What is  ? Its a vector of some kind? Wht is the dmension of this vector? What are its vector components?

This introduces the norm of a vector. What norm is this? The l-2 norm? some other norm? (See Lp space for a discussion of vector norms).

The next section says:

Given the time series x, and some real number r > 0, for every pair i, j where i is one point in time, j is another point in time, if then points x(i) and x(j) are isometric.

In this paragraph, x seem to not be a vector. How is this related to the previous x? is it the same thing? Something different? Also, this time, the norm is drawn with a single vertical bar, not a double vertical bar. Is this a different norm, or the same norm?

Clarifying these things will go a good ways towards addressing my objections. linas 00:16, 14 February 2006 (UTC)

Why don't you take a tour thru Recurrence plots, and tell me if you understand this better after reading what's there, and if your questions change or remain the same.Lakinekaki

If understanding this article requires all you've just said, then Linas is correct. Also, it would be nice if you learnt how to sign your messages. Also, it would be nice if you avoided pointless spite like [2] that. William M. Connolley 18:35, 14 February 2006 (UTC).

It would be nice if your achievements would match your arrogance. It would also be good for the Earth and everyone living on it, I guess, since you are into the global climate research. I will be signing my messages the way I am doing it all the time, with three ~, or with four ~ that will also include the date. Choice is upon me. If I am not loged in, I will just type my username.

Since you mentioned spite, I'll tell you something about it. People (2-3 of you) are objecting here to the things that are very simmilarly described in other mathematical articles, and I can't find simmilar objections there. Think again about who is being spiteful. Lakinekaki <--- (I will sign it any way I want. If you would give me some tutoring lessons 1 on 1, I would be happy to take few lessons if you won't charge me more than $4/hour - since the subject does not seem very complicated.) —This unsigned comment was added by 70.16.63.151 (talk • contribs) .

[edit] LInk to chaos theory

The only reason I placed the article in Choas theory was that the vote on Chaos theory was against a direct see also link. Hence there is now now root to find Bios from Chaos, even though Bios is squarely a topic in chaos theory, has pear review papers, has a faily well know mathematician on the team, blah blah. Ah well horray for wisdom of the mob. --Salix alba (talk) 17:59, 6 April 2006 (UTC)

What's a pear review paper? In any case, it has one, count them, one peer reviewed paper. — Arthur Rubin | (talk) 18:04, 6 April 2006 (UTC)

OK Peer - I make it six. --Salix alba (talk) 18:14, 6 April 2006 (UTC)

[edit] Quick comment

Present version claims

1. "Bios theory is being applied in mathematics, medicine[9], psychology, economy, etc." Applied in mathematics? Maybe so, but my background is in dynamical systems, yet I have seen no evidence of widespread application of any "bios theory". Please be more specific about who is working on these alleged applications.
2. "Rush University is implementing this theory in medicine and psychology." This DMS; presumably the claim intended is that some 'person at Rush University is using "bios theory" in his research. Clarify and verify or remove, please.

More generally, I think some qualified user should take a hard look at all the claims made in this article, and modify as needed. Since my own diss concerned dynamical systems theory, I am qualified, but I am still ather exhausted from the endless arguing with User:RayTomes over Edward R. Dewey and all his cycle theory (sic) articles. At a glance, this article does appear more sophisticated than the vague numerological mysticism of Dewey/Tomes, but a closer look might reveal otherwise. ---CH 00:50, 11 May 2006 (UTC)

[edit] Suspect claims in the first paragraph

I count at least nine suspicious claims just in the very first paragraph of the current version!

1. The name "bios theory": what makes this a scientific theory? This is not a characterization which should be bestowed lightly.
2. The theory states that causal processes can generate new patterns, not only random or stochastic processes. I guess that by "causal process" the author means "deterministic process"? If so, how does "bios theory" differ from plain vanilla chaos theory, which is often motivated by pointing out that something like the Mandlebrot set is produced by a simple deterministic process? Or the full two-shift, which is a deterministic process but which exhibits "random" behavior in various senses? (These senses include notions of randomness from ergodic theory and from algorithmic information theory.)
3. "Biotic patterns are found in ... the distribution of galaxies along the z (time-space) axis". Is the author claiming that galaxies are alive? Or simply that the apparent spiral pattern is due to a deterministic process resembling some biotic process? If so, what deterministic process? (Kepler's laws, I hope.) And what biotic process? Resembling how? Is the author aware of an extensive body of work in astrophysics on the apparent spiral patterns? (Hint: the word "apparent" is significant.)
4. "Biotic patterns are aperiodic cycles" In dynamical systems theory, cycles are generally periodic by definition (see textbooks on symbolic dynamics). Did the author have in mind a more general term such as oscillation, or even resurrence as in the Poincaré recurrence theorem and Kac recurrence theorem in ergodic theory?
5. generation of "diversification, novelty, asymmetry and hence lower entropy" This seems to reveal many serious misconceptions about entropy. First, in information theory and in the measure-theoretic formulation of the notion of a dynamical system, greater novelty is in fact associated with a larger Shannon entropy or Kolmogorov-Sinai entropy respectively. These are very closely related notions (see Peter Walters, Ergodic theory, Springer, 1981 or this webpage (which I wrote long long ago; this is an unauthorized archive with many broken links, unfortunately). Kolmogorov-Sinai entropy is closely related to topological entropy and Shannon entropy is closely related to a combinatorial notion, Boltzmann entropy. Indeed, many authors motivate Shannon's definition of his entropy in terms of "surprise value". Second, one can easily reformulate and greatly generalize Boltzmann's definition using the notion of a suitable group action to define the combinatorial entropy of a subset as the log of the index of the pointwise stabilizer subgroup, as defined in elementary Galois theory. So greater asymmetry is associated with a smaller stabilizer and therefore a larger index and thus a larger Galois entropy.
6. "causation" Shannon entropies can be regarded as a nonparametric measure of correlation (bearing a precise relation, in some contexts, with Chi-square), but one cannot make inferences about causation from a positive interaction entropy (written I(X,Y) in Shannon 1948 and most subsequent work).
7. "bounded (e.g. heartbeat intervals, shorelines)" Further clarification is needed here.
8. "high frequency chaotic oscillations" Likewise questionable.
9. "Many aperiodic processes suspected to be the product of chance or the signature of natural chaos appear to be biotic." The first half of this sentence suggests that the author may be missing one of the major lessons of ergodic theory, combinatorics, and graph theory, namely that "random" patterns exhibit a very strict kind of statistic regularity (c.f. the Shannon equipartition theorem as in Thomas & Cover, Elementary Information Theory, Wiley, 1991, and allied results such as the Szemerédi regularity lemma; see also random graph). The second half again suggests he may believe that various phenomena which are generally not considered to be associated with life are in some sense living. This would be a very extraordinary claim demanding very extraordinary proof.
10. "Bios seems to be a generic, widespread pattern of natural processes, extending from physics to psychobiology." Need I really point out that this kind of claim should be qualified? E.g. According to Sabelli, so called bios represents... and so forth.
11. "Bios is generated mathematically by feedback processes" Is this a definition, a theorem, a conjecture, what?

Good grief, I was going to start reading the rest of the article, but I shrink from the prospect of listing a comparable number of challenges to every damn sentence in this article.

I have added cleanup and POV flags for reasons which should be perfectly obvious.---CH 03:02, 12 May 2006 (UTC)

I don't really want to get into this, but readers should also be aware that while there is a large literature on topics such as "recurrence plots" and "embedding dimesions", these and allied notions are regarded with considerable suspicion by hard analysts trained in ergodic theory. It is very important to understand that not all of chaos theory is of equal quality, still less of equal rigor. I know several knowledgeable pure and applied mathematicians who regard a good deal of so-called applied chaos theory as something akin to snake oil. Even worse are claims by adherents of transparently cranky notions like so-called "bible codes" that such notions are validated by dynamical systems theory, ergodic theory, or what have you. Suggestion for industrious Wikipedian: examine book reviews of books on applied chaos theory in some of the higher quality journals, MR reviews, and so forth. Strongly worded expressions of doubt should not be terribly hard to find.---CH 03:26, 12 May 2006 (UTC)

1.The name "bios theory": what makes this a scientific theory? This is not a characterization which should be bestowed lightly.

Did you actually read the scientific theory? - Theory has a number of distinct meanings in different fields of knowledge, depending on the context and their methodologies. In common usage, people use the word "theory" to signify "conjecture", "speculation", or "opinion." In this sense, "theories" are opposed to "facts" — parts of the world, or claims about the world, that are real or true regardless of what people think....There is also scientific method involved:

Define the question

Gather information and resources

Form hypothesis

Perform experiment and collect data

Analyze data

Interpret data and draw conclusions that serve as a starting point for new hypotheses

Publish results

Could you tell me which step was missed?

2. I guess author had in mind novelty and not just complexity. Novelty as defined by the same author.

3. Is the author claiming that galaxies are alive? No.

Or simply that the apparent spiral pattern is due to a deterministic process resembling some biotic process? Nobody is talking about the patterns of individual galaxies, but about the distribution of galaxies. When distributions of galaxies are analyzed, each galaxy is considered a point in space, without taking into consideration their forms.

4. Did the author have in mind a more general term such as oscillation... I don't know what the author had in mind, I know what he wrote.

5. I think that what you wrote will be enlightening to the author.

6. Agree. There should be indicate, not demonstrate.

7. I don't think that I can find that clarification in papers - only a weak definition.

8. I agree, but again, that's what I found in papers.

9. I don't understand why you talk about life again here.

10. Of course! At the beginning of paragraph, it sais, Sabelli et al.'s theory attempts to characterise... You can repeat it here too.

11. Conjecture and experimental finding.

3 more things: 1)you are right that article needs improvements and qualifiers (e.g. according to Sabelli..). Feel free to add them.

2)you are doing a lots of OR here. Article should present what has been published. If you don't like it, that's one thing. WP:VERIFY - The threshold for inclusion in Wikipedia is verifiability, not truth. This means that we only publish material that is verifiable with reference to reliable, published sources.

3)biotic does not mean alive, as you are interpreting it. it means systems that generate novelty, etc.Lakinekaki 12:52, 12 May 2006 (UTC)

[edit] Apparent conflict of interest by the author of this article

User:Lakinekaki uploaded the images for the figures to Wikimedia. Lakinekaki cited the Chicago Center for Creative Development as the source of the images. This website is registered to one Linnea Carlson-Sabelli of Chicago, IL, whose email address suggests some affiliation with the Rush University Medical Center. It is natural to suspect that this "CCCD" is the unspecified affiliate of "Rush" which is mentioned in the article. (However, I suspect that the implied affiliation of "bios theory" with Rush Medical Center is greatly exaggerated.)

Carlson-Sabelli is apparently one coauthor of a paper

Sabelli, H.; Sugerman, A.; Kovacevic, L.; Kauffman, L.; Carlson-Sabelli, L.; Patel, M.; & Konecki, J. (2005). "Bios Data Analyzer". Nonlinear Dynamics, Psychology and the Life Sciences 9: 505-538. 

which is cited in the article. It is natural to guess that there may be some relation between Linnea Carlson-Sabelli and Hector Sabelli, who is described in the article as a medical doctor and as the author of "bios theory" (sic).

I note that Lakinekaki has also edited Chaos Data Analyzer. Is this connected with the Bios Data Analyzer in the paper just cited?

The CCCD website claims "sponsorship" by something called the Society for the Advancement of Clinical Philosophy. It appears from this link that the Society for the Advancement of Clinical Philosophy has also provided financial support for another paper, by L. Kauffman, M. Patel, J. Konecki and Linnea Carlson-Sabelli.

One file available at that website, The Infinite Attractor of Evolution, claims "This provides a mathematical metaphor for God compatible. with contemporary science". This suggests that the CCCD has an extrascientific agenda which calls into doubt whether it can be regarded as constituting a scientific organization at all.

My concern is heightened because at this page from the CCCD website, one Lazar Kovacevic is listed as a research associate, and he is said to have earned a BSEE from the University of Belgrade. I noted that on his user page, Lakinekaki says he now lives in Chicago but has previously lived in Belgrade, and he also mentions a foreign engineering degree. On his talk page he seems to imply that his website is Explore Ideas, which is registered to Lazar Kovacevic of Chicago, IL, whose email handle is indeed "lakinekaki". Note that Kovacevic appears to be a coauthor, with L. Carlson-Sabelli and H. Sabelli of the paper mentioned above.

Examining the history page shows that this article has been created almost entirely by User:Lakinekaki and the ameritech.com anon, who has used the IP addies 65.42.82.144, 65.42.94.166, 65.42.84.184 (ameritech.com, all registered to Southwestern Bell Internet Services; geolocated in Chicago, IL) and 69.33.60.41 (megapath.net, aka AMERITECH.NET; geolocated in Chicago, IL).

So it seems that this article is almost entirely the creation of someone who is apparently employed by an organization (Chicago Center for Creative Development, with financial support provided by this Society for the Advancement of Clinical Philosophy) which exists solely in order to promote the ideas of H. Sabelli, an organization whose scientific credibility appears to be in grave doubt. It also appears that Lakinekaki has been concealing from readers of the WP a hidden agenda which has nothing to do with science. ---CH 06:25, 12 May 2006 (UTC)

[Edit conflict; look back a version, if neccessary]. "This provides a mathematical metaphor for God compatible with contemporary science" Makes a good Templeton Prize contender entry though. Good web snooping, Hillman, but you overstep your facts. You find that BiosTheory, (maybe better called HectorSabelliism) is connected institutionally with Rush Medical Center, which is thus "dubious", but you present nothing dubious about Rush other than its promotion of Bios theory. Sorry, not quite logical. You need to establish Rush's dubiousity in some independent way, before you can use it as an argument (by association) against Bios Theory. Anyway, that's not going to change anything about what happens on Wikipedia. I say, "Let a thousand flowers bloom", but where? It seems HectorSabelliism is part of what is a comprehensive philosophical system, (and not yet a science ((pending further reasech by yours truly))) and surely we have articles about comprehensive philosophical systems on wikipedia. But is it a "notable" one? Not clear that it is. Of course, if Hector Sabelli were to get involved in a major public scandal, then this article would definitely be suitable for Wikipedia. Anyway, I note that there is a similiar article at http://wikinfo.org/wiki.php?title=Bios_theory , but lacking the images. --GangofOne 06:47, 12 May 2006 (UTC)

GoO, I see now that I wasn't sufficiently clear, so I just modified my comment above to clarify this point. I do not suspect that Rush Medical Center appears to have a hidden agenda; rather, I suspect that the Chicago Center for Creative Development, which is financially supported by this Society for the Advancement of Clinical Philosophy, appears to have a hidden agenda. My best guess is that Linnea Carlson-Sabelli (and maybe also Hector Sabelli) happens to work at Rush Medical Center, but that "bios theory" (sic) has in fact very little connection with Rush Medical Center. Just to emphasize: I currently see no reason to suspect that Rush Medical Center is a house of quackery or anything like that. I know very well that most hospitals are huge organizations which employ all kinds of people, including doctors who sometimes have some very odd ideas but who may be perfectly competent as physicians.

It seems to me that the article fails to clearly present "bios theory" as a "comprehensive philosophical system" rather than as a scientific theory. The cited papers and the article in fact seem to suggest that the authors view it as part of dynamical systems theory/chaos theory. ---CH 07:02, 12 May 2006 (UTC)

It would be nice to just delete the whole thing as gobbledegook... though the new header does help William M. Connolley 07:13, 12 May 2006 (UTC)

I have also been considering initiating an AfD. Googling on "Bios theory" and "Sabelli" returns only five hits, two of which are this article and a mirror. ---CH 07:18, 12 May 2006 (UTC)

Thanks for clarifying your comments, Hillman. As far as "comprehensive philosophical system" goes, I was thinking of the book you linked to above, Bios: A study of Creation Hector Sabelli, et al. Bios theory seems to be only one part of the thoughts in this work (based on what I see on the web page [3]) (seems like these guys should get in touch with Stephen Wolfram and his New Kind of Science .... ) --GangofOne 07:58, 12 May 2006 (UTC)

I was one of several authors in discussion with Lakinekaki when he first wrote this article and tried to promote its importance by giving it a prominent link from chaos theory. FWIW, my views are (i) there is nothing in "bios theory" or "biotic behaviour" that makes it qualitatively different from standard chaos theory; (ii) it is okay to have an article on bios theory, which is after all a published theory, but the article should make it clear that very few mathematicians or scientists believe that bios theory describes anything especially new or universal; (iii) Lakinekaki should clarify any link he has with the authors of the bios theory papers. Gandalf61 08:16, 12 May 2006 (UTC)

I am a programmer. When I do some programming related to bios, my name gets into related paper. This means that I don't make theories. I write code that uses definitions from theory, and analyses the data. I think that this theory has some good ideas, and that's why I wrote an article about it. There are so many articles on Wikipedia, and rarely do I look for something that is already not here. When I find missing article, I add it. I happen to know about bios theory, and that's why I wrote about it. I also wrote about few other things that were missing. Lakinekaki 16:31, 12 May 2006 (UTC)

It seems that related article by User:Lakinekaki has already been deleted following an AfD for Bios Data Analyzer (as in the title of the paper coauthored by Kovacevic and mentioned above). Yes, this stuff seems to have much in common with other highly dubious claims involving chaotic time series.

On behalf of a pediatric cardiologist, I once critiqued something called approximate entropy, which was introduced by one Steven Pincus in the context of cardiology (see Lakinekaki's mention of similar motivation for his computer program). Now this is really serious stuff because pediatric cardiologists are rather desperate for a reliable way to predict the onset of blood poisoning in premature infants in time to save the infant. There are many possible indicators one could investigate, but cardiologists naturally tend to think first of extracting subtle signals from EKGs. I am particularly concerned by the potential for loss of life should anyone actually try to use something based on such dubious "science" as "bios theory" (sic) in a hospital, since I fear that this could very easily amount to a physician lying to himself with statistics. In the U.S. alone potentially thousands of very young lives are at stake each year.

GoO, I have also criticized the even more widely publicized ideas of Roy Frieden and Wolfram. One of the members of my thesis committee was briefly interested in cellular automata a long long time ago, but I have the impression he was soon disillusioned. I don't know anyone with a background in symbolic dynamics who has been favorably impressed by A New Kind of Science.

Gandalf61, while I agree that Bios theory seems to attribute to "bios theory" (sic) ideas which are more properly attributed to Chaos theory, I believe that the truly "novel" bits of "bios theory", such as this alleged measure of "novelty", are seriously flawed. On the face of it, it seems clear that "bios theory" is not notable even as yet another instance of junk science. The only question in my mind is whether the potential for loss of life exists. Is this stuff actually being used to treat patients in the ICU? I hope not! ---CH 08:25, 12 May 2006 (UTC)

This is worrisome, just noticed specific reference to heart rate intervals, which suggests that indeed Sabelli might be trying to apply this stuff to treat sick infants. To clarify the situation: every year in the U.S. alone thousands of premature infants wind up in neonatal ICUs. These very ill infants are hooked up to all kinds of shunts and things. Unfortunately, bacteria tend to wick up these things and the infants are at risk of contracting sepsis (blood poisoning). The problem is that, on the one hand, by the time clinical indications of sepsis are unambiguous, mortality is very high (something like 30%, as I recall), and on the other, treating for sepsis when the infant has not in fact developed sepsis is often fatal. So it's a high stakes roll of the dice for the cardiologist who thinks one of his young patients might be in the early stages of developing sepsis. Now there is apparently some tantalizing evidence that the onset of sepsis might be indicated in these patients by subtle changes in the heart rate inervals in time to reliably diagnose sepsis and to intervene clinically. Hence the desperate search for a better way to extract data from time series derived from EKG readouts. I suppose one could argue that even a marginally effective technique is better than guessing in the dark, but I must say that I find the prospect of using such dubious "protoscience" to make critical care decisions in the treatment of infants utterly horrifying. I certainly hope that "bios theory" is not being "applied" in this way.---CH 08:52, 12 May 2006 (UTC)

FYI. paper cited http://creativebios.com/webjass/1EmpiricalFoundationsAndMedicalApplication.pdf (much on heart rate variations in this paper) references papers by some of same people referenced in "approx. entropy-Pincus" paper: Goldberger AL, Mietus J, Lipsitz LA. (Sabelli is MD.) On other point, I mention Wolfram because of thematic commonality with the Bios verbiage, "novelty" out of chaos, supposed profundity, etc. Maybe if they got together they could actually come up with something; not that they already have it. --GangofOne 09:38, 12 May 2006 (UTC)

Yeah. Good grief... maybe we should email some cardiologists to inquire? ---CH 02:13, 13 May 2006 (UTC)

Written by a notable person [4] here you can read: There is preliminary evidence to suggest that injecting variability that emulates ordinary 1/f fluctuations into physiologic supports can accelerate recovery, providing improved outcomes at reduced costs. Bios theory talks about variability, novelty, continuity, etc. Bios theory also talks specifically about heart rate variability in healthy hearts, and about absence of the same in unhealty hearts... Lakinekaki 12:38, 12 May 2006 (UTC)

Given the fact that you have been caught in what I regard as a major deception, I think you have an extraordinary burden to come clean about the affiliations/nature of organizations/people you might cite from this time henceforth. You linked to files at two websites,

1. how-why.com, registered to the Howhy(TM) Company of Urbana, IL,
2. sccm.org, registered to Society of Critical Care Medicine of Des Plaines, IL,

without explaining what this company/org sells/does or whether you, Buchman, or Sabelli have any connection to them. For the sake of argument, however, let's pretend that you have established that Buchman is notable as a past president of SCCM. Well, the quote you gave has nothing to do with "bios theory" (sic) but in fact just restates part of the rationale for the "desperate search" I mentioned above in connection with pediatric cardiology, which predates the advent of "approximate entropy", which in turn predates "bios theory"! I don't know who you think you are fooling, User:Lakinekaki.---CH 02:13, 13 May 2006 (UTC)

For what it's worth, I'm now almost sure this article doesn't belong here on wikipedia, but there was a serious, scientific paper based on catastrophe theory supporting adding "noise" to pacemakers decreased the probability of fibrillation. I doubt that bios theory provides any better basis. (Or that the referenced paper by Buchman deserves any weight if it doesn't reference that paper.) — Arthur Rubin | (talk) 04:12, 13 May 2006 (UTC)