Talk:Extreme physical information

The controversial claims of Roy Frieden
This article and other B. Roy Frieden related articles needs to be fact checked and POV-pushing edits need to be ameliorated. In particular, edits by the data.optics.arizona.edu anon raise the issue of conflict of interest re WP:VAIN-WP:NPOV-WP:AUTO, since this anon is IRL none other than Roy Frieden (who is Prof. Em. of Optical Sciences at the University of Arizona). The data.optics.arizona.edu anon has used the following IPs to make a number of questionable edits: Friedenr 21:46, 26 June 2006 (UTC) Comment: How "misleading"? A. Stam (Dutch mathematician) had previously proved that it can in fact be derived from the Cramer-Rao inequality. The information aspect of that derivation is covered in the article's main ref. Science from Fisher Inf. pgs. 153-157. Friedenr 21:46, 26 June 2006 (UTC) How "controversial"? It is in fact derived on pgs. 221-227 of Science from Fisher Info. book. Also, why the quotes around "method"? Yes, EPI is a method, also a world view (related to Wheeler's), a computational procedure, and a measurement process. Friedenr 21:46, 26 June 2006 (UTC)It is indeed derived (again, without quotes) on pgs. 131-139 of Science from Fisher Inf. book Friedenr 21:46, 26 June 2006 (UTC)highly dubious ? I refer the writer to the many peer-reviewed and published applications of EPI to physics in journals. The scope is broad, meriting a general approach description.
 * 1) 16 June 2006 states controversial claims as established fact in Extreme physical information
 * 2) 16 June 2006 adds "Physics actually derives out of information, through the principle of Extreme physical information" to Information; this is egregiously misleading, in my view
 * 3) 6 June 2006: adds cites of his papers to Extreme physical information
 * 4) 23 May 2006 adds uncritical description of his own work in Lagrangian and uncritically cites his own controversial book
 * 5) 20 May 2005 confesses IRL identity
 * 6) 22 October 2004 attributes the uncertainty principle to the Cramer-Rao inequality, which is potentially misleading
 * 1) 22 October 2004 attributes the uncertainty principle to the Cramer-Rao inequality, which is potentially misleading
 * 1) 21 October 2004 adds uncritical mention of his controversial claim that the Maxwell-Boltzmann distribution can be obtained via his "method"
 * 1) 21 October 2004 adds uncritical mention of his controversial claim that the Klein-Gordon equation can be "derived" via his "method"
 * 1) 9 September 2004 adds uncritical description of his work to Fisher information
 * 2) 8 September 2004 adds uncritical description of his highly dubious claim that EPI is a general approach to physics to Physical information
 * 1) 8 September 2004 adds uncritical description of his highly dubious claim that EPI is a general approach to physics to Physical information
 * 1) 16 August 2004 confesses IRL identity
 * 2) 13 August 2004 creates uncritical account of his work in new article, Extreme physical information
 * 3) 11 August 2004 creates his own wikibiostub, B Roy Frieden

It is important that this and other articles should state that while Frieden's work has attracted much interest, it remains highly controversial: Friedenr 21:46, 26 June 2006 (UTC) The phrase attracted much interest is a mild concession to reality; in fact it is widely used. To call it "highly controversial" is to therefore ignore reality. I somehow doubt that its users would agree with the extreme nature of this comment. ''The book has the strong point of deriving most of the Lagrangians of physics, rather than postulating them. The idea of a unifying principle based on measurements and information is appealing... Moreover the EPI principle also provides a framework for forming new Lagrangians. The ideas in Frieden's book will probably encounter some resistance'' [rather prophetic] Therefore, Frieden's uncritical description here of his own work poses a problem for WP. ---CH 20:59, 16 June 2006 (UTC)
 * (the reviewer has some positive comments but concludes that Frieden's work is "misguided")
 * eprint (highly favorable)
 * unauthorized electronic reprint The review is highly favorable, but is not available for free viewing at the New Scientist website. However, it can be read at the preceding site.
 * Physics from Fisher Information: A Unification (a review) from Cosma Shalizi (Computer Science, University of Michigan) (highly critical)
 * Physics from Fisher Information (a review) from R. F. Streater (Mathematics, Kings College, London) (highly critical)
 * Physics from Fisher Information thread from sci.physics.research, May 1999 (non-specialists web discussion, no conclusions)
 * Fisher Information - Frieden unification Of Physics thread from sci.physics.research, October 1999 (miscellaneous web discussions, no conclusions)

Response of B. Roy Frieden
The primary aim of the article "Extreme physical information" (EPI) is to describe a computational approach to science. Its aim is NOT primarily to describe or promote the books that have been written about it. It is important not to confuse the two, as although the books (actually, one book, and NOT the book referenced; see below) have been criticized, the EPI approach itself udeniably works. Witness the large number of people who have published using it (among them several I have now added to the article for clarification). That it works is therefore not "controversial." The above book might be regarded as "controversial," but that's another matter.

Therefore, I request that you delete the disclaimer you have added to the top of the article, that "The neutrality and factual accuracy of this article are disputed." All this does in actuality is to discourage people from using the approach. People in physics as in any other field are timid about associating with a "controversial" approach, both because it might be wrong and because it might compromise their reputation. In this case the approach is decidedly not wrong. If it were, hundreds of papers in highest-level journals (Phys. Rev. A and E, J. Theor. Biol., J. Cancer Research, J. Chem. Physics, Phys. Lett. A, etc.) would be wrong as well. Do you want to claim this?

Now regarding the book itself. Note that ALL the reviews you have listed in the above "discussion" are of the 1st edition of the book, "Physics from Fisher Information." In fact, that first book was corrected, where the criticism was valid (4 places), and also enlarged to include science in general. The result is a second book "Science from Fisher Information." THIS in fact is the book referenced in the article. To my knowledge, this corrected book has only been reviewed by one person, S. Sherif, U.K. (I think Univ. of Kent), and positively; of course you can look it up on google.

I cannot accept an article with the current disclaimer at its top. If you insist on leaving it in, I will just drop the article. The EPI approach - a valid and valuable tool of system analysis - does not need that kind of negative publicity. If you are willing to drop the disclaimer subject to some compromise, such as e.g. not mentioning any book of mine as a reference, I can consider that. It's up to you.

Sincerely, Roy Frieden

13:19, 20 June 2006


 * Too bad (specifically in regard the last paragraph). You do not have control of the article.  The disclaimer can be argued, although, I still think the theory can be interpreted to mean whatever the reader wants it to mean, making it essentially void.  You may add comments to the effect that the negative reviews are of the first edition, and that the second edition is significantly different -- provided that the second edition is in print.  If it's not published yet, you are the only source of the information.
 * However, if you insist that the article cannot be here without your consent, you're wrong. If you attempt to destroy whatever value the article may have, I may argue in favor of keeping it.  &mdash; Arthur Rubin |  (talk) 21:41, 20 June 2006 (UTC)

(Reply by Frieden reformatted by Chris Hillman as per Arthur Rubin's request)

By the way, most people seem to use the version where the invariance constraint is replaced by input constraints. This is an approximate use of EPI, and neatly corresponds to the "empirical level" of knowledge delineated by the philosopher Charles Peirce (circa 1890) in forming a physical theory. His other two levels of prior knowledge exactly correspond to invariance approaches (a) and (b) of EPI (see Peirce in Frieden_Fisher information website mentioned above). --- 15:40, 20 June 2006
 * 1) "Too bad (specifically in regard the last paragraph). You do not have control of the article.  The disclaimer can be argued, although, I still think the theory can be interpreted to mean whatever the reader wants it to mean, making it essentially void." Arthur, please clarify this point, as it comes out of the blue and may have a strong bearing on the matter.  The theory is carefully delineated in pgs. 82-88 (Secs. 3.4.1-3.4.6 of the "Science from FI" book).  This cannot be interpreted to mean whatever the reader wants it to mean.  True, it has flexibility in the invariance principle that has to be decided upon.  Is this what you are referring to?  To connect with the real world, any mathematical approach to reality has to have a mechanism for a physical input.  EPI's mechanism is an invariance principle.  See further on this in my last comment below.
 * 2) "You may add comments to the effect that the negative reviews are of the first edition, and that the second edition is significantly different -- provided that the second edition is in print. If it's not published yet, you are the only source of the information."  That second edition is the very source cited in the article.  It was published in 2004 and is certainly still in print.
 * 3) "However, if you insist that the article cannot be here without your consent, you're wrong. If you attempt to destroy whatever value the article may have, I may argue in favor of keeping it."  Well, this is flattering albeit in an obtuse way.  It's valuable enough to include, but not valuable enough to accept fully.  As I have to keep saying, the approach itself has been used, is being used, and will be used, because it works.  There is no question as to its utility and therefore its validity.  Therefore I have to protest your comment.  Whom else may I contact on this matter?


 * Hi, Roy, and thank you for your prompt. First let me point out that Wikipedia has several processes for disputes over content which work pretty well if everyone remains calm.  Note that talk page discussions are the recommended first step, and I am cautiously optimistic that we can reach a satisfactory result here.
 * "the books (actually, one book, and NOT the book referenced; see below) have been criticized" Can you clarify this, please?  Are you saying that the book by yourself, B. Roy Frieden (2004) Science from Fisher Information, 2nd ed. Cambridge University Press, is not the book which you think is relevant to this discussion?

BR Frieden response This second edition, titled "Science from Fisher Inf.," is very relevant. It is the book I use in the article as its main reference. However, as I have stated above a couple of times, the book that was REVIEWED by everyone (except for S. Sherif) was the FIRST edition, titled "Physics [not Science] from Fisher Information." This book is NOT even one of the references for the article! As I mentioned above, it contained 4 algebraic errors (as cited in the Lavis & Streater review), which were corrected and then included in the SECOND edition (the book that IS cited in the article). By the way, these error corrections were very minor and easily facilitated and in no way compromised the derivations they were in. These remained valid. Thus, the SECOND edition no longer contains these errors.


 * "EPI approach itself udeniably works." That is precisely one of your claims which has been challenged. That is, as I understand it, you claim to have provided a method by which one can derive from an extremum principle the "correct" Lagrangian, in situations where (as often happens) there are many plausible looking choices which a physicist might write down.  But some of the critics has stated (see the links in my comment above to the sci.physics.research threads and to the websites of Cosma Shalizi) that this "method" does not appear to be well-defined, and in fact that it comes down to ad hoc arguments, which if true would obviate the goal, which as I understand it is to replace ad hoc arguments in favor of some choice of Lagrangian by a method guaranteed to give the correct choice.

 BR Frieden response Despite what Shalizi supposes, the aim is NOT to come up with a method guaranteed to give the correct choice in each scenario. There is no such method - nature is just not that simple. I don't know where he got that supposition from. It's not stated as such in the book (note: 2nd edition, not first, as discussed below). For that matter, it's not in the 1st edition either. From either book, the opening section is "0.1 Aim of the book." The first sentence is, "The overall aim of this book is to develop a theory of measurement that incorporates the observer into the phenomenon under measurement." The section then goes on to describe secondary aims, mainly to place all physical law under one general umbrella -- that of Fisher information. Nowhere is it claimed that one fixed, mechanistic procedure will fit all scenarios. Rather, a "structure" is alluded to, which all will follow. That structure is general, with slots (mainly for information J) to fill in. How Shalizi did not understand this is beyond me. I have to state that it is easy to criticize a book when you set impossible goals for it. This in my view is the real problem here.

But returning to the question at hand, of how any new effect is treated by EPI, I responded briefly to it above in my conversation with D. Rubin, but let me respond more fully here. All references are from the book that you do not yet have, "Sci. from Fisher I." To connect with the real world, any mathematical approach to reality has to have a mechanism for a physical input. EPI's mechanism is a three-tiered level of inputs, representing three levels of accuracy for the given problem. These are discussed on pgs. 82-88 of the book. The top level [I call it (b), not (a) merely for historical reasons] is the use of a unitary space to data space. This has to be known as prior knowledge, sometimes called "physical insight" or "intuition". Here each of the two EPI principles - the extremum principle and the EPI zero principle - can give different answers. An example is quantum mechanics (p.133 of the book), where the unitary transform is the Fourier transform from 4-coord. space to 4-momentum space. Here the extremum condition gives the answer for bosons - the Klein-Gordon eq.- and the zero condition gives the answer for fermions - the Dirac eq. This is a beautiful result since in general (by the standard model) there are no other type particles than bosons or fermions.

In general, these answers are exactly correct if the correct unitary space is used. So, what if the "wrong" invariance principle is used? The answer is that it may be wrong for the given scenario, but will be right for some other scenario. The working premise of EPI, which is borne out by experience, is that every reasonable invariance principle when used in EPI gives the answer for some physical scenrio. This has so far been found to be true in all cases tested out. Basically, the user's task is then to match up his measurement scenario with the right invariance principle. As other invariance examples (pages cited are in the book "Science from Fisher Inf.": Rotation of 4-coordinate space by an imaginary angle gives rise via EPI to the Lorentz transformation (pgs. 99-100); Fourier transform from x-functional space to momentum functional (pgs. 296-298) gives via EPI the Wheeler-DeWitt eq. of quantum gravity (p. 299).

The second level down [level (a)] of accuracy is use of the self-consistent approach, where the two facets of EPI - the extremum principle and the zero principle - are solved simultaneously for the amplitude function. Now there is but one solution for the amplitude function (as in statistical mechanics or the Einstein equation of classical gravitation). The solution is quite accurate but not 100% so because it isn't covariant. The third level down [level (c)] is abandonment of knowledge of invariance in favor of inputting empirical knowledge such as data. This gives rise to solutions to problems of econophysics (p. 338), the 1/f power spectral law (p. 258) and Fisher-based laws of statistical mechanics (see papers by Plastino et al in paper listing).

This variety of solutions according to what is assumed as the invariance principle is a strength of the EPI approach, and is why it in fact popular with so many users (see list of users and websites). When confronted with an unknown system, in practice a scientist/engineer wants a variety of possible answers to test against, not rigidly one, since there certainly is NO approach that can get the answer mechanistically, without any experimentation. Nature is just not that simple. Therefore to expect a mechanistic answer is an unrealistic expectation, although it apparently is what Shilazi wants to assume that the book claims.


 * "Witness the large number of people who have published using it (among them several I have now added to the article for clarification)." This is not a useful response to the criticisms of the alleged method which I cited.

BR Frieden response Perhaps then my preceding remarks will help. I might add, though, that the "proof" of any scientific theory is its ability to verified or refuted through use (Popper's well known criterion of "refutability") Thus, by Popper's criterion, all these uses and users I cited are independent verifications of the approach. In fact, according to Popper there is no other way, ultimately, to "prove" the validity of a theory.


 * "All this does in actuality is to discourage people from using the approach." Well, at least one of the above mentioned critics tried to apply the "method" and concluded that in fact there is no method!

BR Frieden response I can make no excuses for someone else's failures. You give me an impossible task of explanation here. The approach is well defined as I have summarized above and as exists in the book. Many people have successfully used it (as per my lists of users). If he could not use it, yet many people have, then I have to conclude it is not the fault of the approach.

I am willing to take a second look at that, but for now the important point is that the existence of numerous critical comments shows that the Wikipedia article needs to be rewritten to respect WP:NPOV. As to whether you or your critics are more correct about the ultimate outcome, see Wikipedia is not a crystal ball. Regarding "discourage people from using", see WP:AGF and my unfinished notes for an essay on Wikipedia quality control.
 * "Note that ALL the reviews you have listed in the above "discussion" are of the 1st edition of the book, "Physics from Fisher Information." In fact, that first book was corrected, where the criticism was valid (4 places), and also enlarged to include science in general." I was in fact not aware of the existence of a second edition (second book?!).  I will try to obtain and study this, but you may need to be patient until then, since both local library copies appear to be checked out.  Please let us known if the book you are referring to is not this:

Response by BR Frieden I do not doubt that both copies are checked out. The method is very popular. As you requested, that is indeed the book I am referring to.

"I cannot accept an article with the current disclaimer at its top."'' Regarding Arthur's comment that you do not have control of this article, see WP:OWN. My hope is that we can agree upon a revision and remove the flag. As I just said, it may take some time for me to obtain a copy of your latest book.

Response by BR Frieden I hope so as well. Let us disseminate knowledge and advance science. Meanwhile I notice that such topics in Wikipedia as "cold fusion" and "intelligent design" are not so flagged. Is then this article on EPI more unreliable and questionable than these highly-unlikely theories? I see as well that the following items are flagged: "Heaven's Stairway", describing a cannabis site, the "Hebrew Gospel of Matthew," a contested religious text, and "Best Buy", a store chain of questionable value. Is this the appropriate company for EPI? This is why I want it unflagged.


 * "The EPI approach - a valid and valuable tool of system analysis - does not need that kind of negative publicity." Roy, of course you believe in the validity and value of your work, but your primary responsibility as a WP editor is to serve the readership of the Wikipedia by helping us to provide unbiased and factually accurate accounts of current knowledge.

Aside from all the users of EPI, you can contact the following Ph. D. physicists/engineers for statements on the validity of the approach: A. Plastino (plastino@sinectis.com.ar ), P.W. Hawkes (hawkes@cemes.fr ), Ray Hawkins (Raymond_Hawkins@Countrywide.Com ) , Stephen Ames (sames@unimelb.edu.au ) , H. Cabezas (Cabezas.Heriberto@epamail.epa.gov ), S. Luo (luosl@mail.amt.ac.cn), R. Venkatesan (r_c_v@vsnl.com )
 * "The theory is carefully delineated in pgs. 82-88 (Secs. 3.4.1-3.4.6 of the "Science from FI" book)." I've only seen the earlier book, so we will have to wait for me to obtain a copy so that I can examine this section.
 * Again, I am cautiously optimistic that we can resolve this by discussion on this talk page, together with suitable modification of the article itself.
 * By the way, I think it would be helpful if you registered a user account (note that you do not need to provide an email address in order to register) with a transparent user name like "Frieden".---CH 01:59, 21 June 2006 (UTC)

Resonse by BR Frieden Can you itemize exactly how I set up such an account? Is it free? Thanks.

please refactor
Our Wikipedia software doesn't reflect line breaks. I'd appreciate it if you would refactor using ":" or "*" at the beginning of each (logical) line so that it will work properly. I'll reply to the content later, but you're making it difficult to read. &mdash; Arthur Rubin | (talk) 22:59, 20 June 2006 (UTC)


 * I think Frieden is too inexperienced here to do this himself, but I have just done this for him. I am also adding a header above which I think is useful for newbies. ---CH 01:25, 21 June 2006 (UTC)


 * I appreciate the refactoring. I probably won't make many substantive comments for a day or so, though.  &mdash; Arthur Rubin |  (talk) 01:38, 21 June 2006 (UTC)

Creating an account
Mr. Frieden, setting up an account is easy and free. You see on the left hand side where it says "Sign in/Create account"? Click it. It will now ask you to sign in and give you the option of creating an account. When you click on "create account" it will simply ask you to choose a username and password. After that you're done! Zyxoas (talk to me - I'll listen) 09:30, 22 June 2006 (UTC)

BR Frieden remarks June 22, 2006  refactored by Arthur Rubin, as the intent was clear

(1) Someone deleted my 7 recently added website references to recent papers (2000-2006) using EPI. Do you know anything about this? These papers are important evidence in support of the validity and utility of EPI. If it was not done by someone at Wikipedia for technical reasons, it amounts to malicious and prejudicial behavior by a user.

(2) I added "Some people find it controversial." at the end of the article, as an attempt to defuse this issue. I hope you can accept this as a solution.

(3) I carefully reread the article in view of your apparent claim that it violates the "neutrality" required of Wikipedia. I am satisfied that it is indeed neutrally written. It is neutral because it is merely a description of a new approach to physics. It makes no claims as to providing an automatic mechanistic route to physics. It merely claims that physics fits under the umbrella of Fisher information. Can that in itself be a biased view or an unfounded claim?

(4) The aim of the "Sci. from Fisher inf." book is to show that the known laws of physics do fit under the Fisher umbrella. So far it has not been reviewed, except for one favorable brief review by Prof. S. Sherif (UK, Univ. of Kent). It apparently has no errors. Indeed the 4 errors pointed out by Lavis & Streater in the previous book "Physics [not Science] from Fisher Inf." were specifically corrected as part of the rewrite of the first book.

(5) Tell me where you see any unproven claims in the article and I'll be glad to seriously consider amending them. (6) I realize you have a problem in that on one hand you have some reviewers who disagree with the approach, and on the other hand you have many modern scientists and engineers who are actually USING it in their research. Which should you give more weight to? The opinions of some "experts" or the experience of many more people who actually use it? I think there is no question as to what the answer should be.

(7) I am very dissatisfied with that negative label at the top of the article. This is for reasons that were laid out in detail in previous letters. I am trying to cooperate with you at this point. If you do not remove it in a reasonable amount of time (the order of weeks at most) I shall have to take a less cooperative attitude and resort to other means.

—The preceding unsigned comment was added by Friedenr (talk • contribs) 19:08, June 22, 2006  (UTC)


 * (1) I think I deleted the references.  It may have been a mistake, though.  (You might verify that the links -- at least one of them was bad, when I tried it a few times.)

Friedenr 22:40, 26 June 2006 (UTC)Thank you for your honesty. I will re-verify the links.
 * (2) I've tried to update the comment. It still needs some editing, and the claim of vacuuity needs to be moved up.

FriedenrWhat do you mean by "claim of vacuuity"? I don't understand your point here.
 * (3) The article is not merely a description of a new approach. The article implies it's a better approach.  That's biased.

Friedenr 22:40, 26 June 2006 (UTC)Okay, a real claim. Then the onus is on you to show evidence for it. Where does the article say or even imply that it's a better approach? I certainly have never had this viewpoint, even implicitly. The aim was to introduce EPI as, rather, an alternative approach. Re-reading it to find evidence for your view comes up with blanks, at least for me. Now, if you show me instances of your view I will be glad to change the language as needed to emphasize the neutrality of the viewpoint.
 * (4) That may be. If it hasn't been reviewed, that suggests it's not ready for Wikipedia, though.  In the case of contraversial science, Protoscience, Fringe science, or Pseudoscience, we're not supposed to take sides, so we're not supposed to provide information about theories which haven't been reviewed.

Friedenr 22:40, 26 June 2006 (UTC) But I said that it had been reviewed, by S. Sherif, and positively. Since you mention the terms Protoscience, Fringe science and Pseudoscience, is this what you want to call EPI? And, if so, is this justified in the face of all its past, present and future applications in reputable journals of physics and engineering?
 * (5) The axioms probably shouldn't be stated here, as there's insufficient background to tell whether they make any sense. I tend to think they do not, but I just don't know.

Friedenr 22:40, 26 June 2006 (UTC)What axioms? My (5) above does not mention axioms. It says:"(5) Tell me where you see any unproven claims in the article and I'll be glad to seriously consider amending them." As for a specific claim: Frieden (2004) employs EPI to derive a number of fundamental laws of physics, as well as some established principles and new laws of biology, the biophysics of cancer growth, chemistry, and economics. should not be there unless independently verified. Friedenr 22:40, 26 June 2006 (UTC)I do not have the Wik. EPI article in front of me, but I do not believe that that passage is mine. It's apparently something written by a reviewer, over whom I have no control.

For example, as you're not an expert in biology, a biologist should be credited with verifying that your "new laws of biology" are "new" and accurate. Friedenr 22:40, 26 June 2006 (UTC)Okay, for example those "new laws of biology" have been published in papers, in J. Theor. Biol. and in J. Cancer Research. They were refereed by biologists.
 * (6) If the "experts" really are experts (or, at least, have expert credentials), I lean toward the experts. In fields I'm familiar with, there are a number of "new approaches" which actually provide fewer and less general results than existing approaches, but are still used.  It's a difficult problem, but we need to include critics of the theory and method if they're at all credible.

Friedenr 22:40, 26 June 2006 (UTC)In fact EPI has a wider scope of application than ordinary action-based Lagrangian theory (altho I do not make this claim in the article). The latter require closed, conserved systems whereas EPI allows open ones -- this is why it obtains Lagrangians for cancer growth (see ref. in article), for population growth (in the book ref.) and for economic growth (again, in book). (Note that all these are again from published papers and passed through professional peer review). You won't find action-based Lagrangians published for these fields.
 * (7) I think it's still biased and has factual problems. Perhaps the field isn't ready for Wikipedia, if you can't find (even off-line) a detailed review of your book.

Friedenr 22:40, 26 June 2006 (UTC)Okay, here you've probably expressed your real view. You personally will just not allow it to be listed as a regular article, no matter what evidence I present. Your mind was made up long ago and I've been wasting my time here. I would like to know if Mr. Hillman or someone else can be used to independently judge as well. Please inform. By the way, would it be possible to view your curriculum vita? This might tell me where you're coming from (and have been). Friedenr 22:40, 26 June 2006 (UTC)If EPI is not "ready for Wikipedia" then I submit that the article should not be in Wikipedia. It's beginning to look like I made a real mistake getting into this website. I deeply regret that. Or is this like the Mafia, where you can't get out? &mdash; Arthur Rubin | (talk) 23:52, 23 June 2006 (UTC)

Axioms

 * Quoting from the article:

Frieden (2004: 81-84) grounds EPI in three axioms:
 * Law of Conservation of information change. Let an act of observation perturb a source and the data gathered from that source. Let the perturbed information be δI and δJ. Then δI = δJ.
 * There exist sequences of functions $$i_{n}(x)$$ and $$j_{n}(x)$$ such that $$I = \int \sum_{n} i_{n}(x)\,dx$$ and $$J = \int \sum_{n} j_{n}(x)\,dx.$$
 * Microscopic zero condition.  $$\exist \kappa, 0<\kappa<1,$$ such that $$i_{n}(x) - \kappa j_{n}(x) = 0.$$

Those axioms do not have sufficient context to be included in the article. If you want to add context, go ahead, but without that context, I don't think they belong here &mdash; whether or not your theory is useful, those axioms are not useful to the reader of the article.

As for my CV, I'm not presently in an academic institution, so we call it a "resume", but I have two published papers in information theory among my 20-odd papers, and have some uncredited work in information theory with Robert McEliece and in set theory with my late mother (and probably some other mentions in journals and books). My Ph.D. at CalTech is (primarily) in mathematical logic, and I satisfied the course requirements for a minor in physics, although I didn't take the qualifying exams required to actually get that minor. (Very little of this paragraph satifies [[WP:V]erifiabilty requirements. I just thought you might want to know who I am.)

I may have more specific critisms of the article later, but your quotes from other reviews should at least be balanced by quotes from the critical review. &mdash; Arthur Rubin | (talk) 00:39, 27 June 2006 (UTC)


 * I could easily add whatever context you want, perhaps e.g. inserting the definition of the i_n (Fisher components).  I am quite willing to compromise, with goal of removing that off-putting heading.  However, I have the impression that no compromise toward this goal is possible here.  Your mind is made up and nothing will change it.  You are giving entirely too much weight to self-serving "reviews," and virtually none to actual empirical evidence for the approach.  Empirical evidence is the ultimate "proof" in physics (see the well known work of C. Popper).  This is compared to math (seeing you are a mathematician), where instead rigorous proof provides the basis for acceptance. Thus, Lavis & Streater in their review bemoaned the lack of "rigor" of the approach, something completely beside the point here.  Their one positive was to actually find algebraic errors, which no one else had done.  These were easy to correct and were incorporated into the 2nd edition (the main ref. for the article).  By the way, the corrections were naturally accomodated by the overall approach.  150.135.248.180 19:31, 27 June 2006 (UTC)


 * "Empirical evidence" is not the ultimate "proof" of a physical theory &mdash; usefulness is.

150.135.248.180 22:45, 28 June 2006 (UTC)I submit that the wealth of modern uses of EPI in the new papers that I have listed is evidence of such usefulness.

A theory can be useful even if the results are wrong, on the other hand an accurate theory which provides no insight is probably not useful. 150.135.248.180 22:45, 28 June 2006 (UTC) Agreed. I agree this is different than Mathematics.


 * I still don't see how the Bactra review and the SIAM review could have been of the same book, though.

150.135.248.180 22:45, 28 June 2006 (UTC) A very apt observation. Unless the Bactra reviewer developed a similar theory, or is otherwise upset with the field, or the SIAM reviewer is a co-worker of yours, there's something seriously wrong. 150.135.248.180 22:45, 28 June 2006 (UTC)The SIAM review was completely impartial. That reviewer (someone named James Case) is someone I have never had contact with. I likewise do not know the Bactra reviewer, nor whether he is upset, as you say. But your observation is correct -- something is seriously wrong here.


 * I admit I'm suspicious of claims of new physical metalaws; but it looks like I may have reason to be. Neutron Spectra seems to actually be minimizing I, rather than I - J; and, for all I know, some of the others may be minimizing "information entropy" rather than "Fisher information" or your I - J.

150.135.248.180 22:45, 28 June 2006 (UTC) They are minimizing I - J where J is their constraint terms. The latter consist of their CT tomographic data. This is a use of EPI at level (c) of accuracy (lowest accuracy, due to use of empirical data instead of either a unitary transformation or an invariance statement such as continuity of flow; see previous on this). Perhaps when the procedings of the 2006 ICCS conference are published, that provide sufficient WP:RS to justify the article. 150.135.248.180 22:45, 28 June 2006 (UTC)I am curious to know whose papers in this proceedings you are interested in. If I'm not mistaken, one or more papers using EPI by R.C. Venkatesan will be there. 150.135.248.180 22:45, 28 June 2006 (UTC)Meanwhile, I direct your attention to about 4 more recent EPI papers listed above under External Links. &mdash; Arthur Rubin | (talk) 22:11, 27 June 2006 (UTC)

.

Citations & References
They are over whelming, and in the positive. Vufors
 * I agree, the work covers Notability, Verifiability and Importance, the maths works. There are No "controversial claims". 220.240.249.134

Removal Of The Tag
CONDITIONS - As per Wiki, the Tag conditions come in two parts:

'''[A]. Neutrality.'''

As the Neutrality is not in dispute (see above), this condition is now regarded to be redundant.


 * PASS the Test. Vufors 08:59, 12 July 2006 (UTC)
 * Neutrality: Only one critical phrase, and one critical review. The critical review, alone, deserves a full paragraph, unless specifically discredited by professionals.  Reinserting Unbalanced in place of POV Arthur Rubin |  (talk) 14:56, 12 July 2006 (UTC)
 * This is only one opinion and using the Wiki rules, needs to be tested on this page. I challange your assertion. See next Header. Vufors 04:44, 13 July 2006 (UTC)

'''[B]. Factual Accuracy.'''

We only have the following Wiki condition to consider.

From the Wiki Conditions for Factual Accuracy, Quote;

The accuracy of an article may be a cause for concern if:

[1]. it contains a lot of unlikely information, without providing references.
 * The references are numerous and listed correctly. Vufors
 * PASS the Test. Vufors 08:52, 12 July 2006 (UTC)

[2]. it contains information which is particularly difficult to verify.
 * Its verifiable by its results when applied mathematically.Vufors 08:52, 12 July 2006 (UTC)
 * PASS the Test. Vufors 08:52, 12 July 2006 (UTC)
 * Your(Vufors's) claim that's it's "verifiable by its results when applied mathematically" is disputed. It's clearly mathematically incorrect, as noted in the references I can see &mdash; it's not invariant with respect to coordinate systems.  Hence, in some coordinate systems, it must be incorrect.  (For spacial data, one could argue that the cartesian coordinate system is appropriate, but for non-spacial data, there's no clearly preferred frame.)  Whether it's physically correct, and/or applies to other subjects, should be left to experts in other subjects.  As the only editors who are familiar with the subject are proponents, this will make it difficult to edit.Arthur Rubin |  (talk) 14:56, 12 July 2006 (UTC)
 * "clearly mathematically incorrect" then prove it incorrect. If the mathermatics were false than it would fail... producing wrong answers... but it does not do this, the operations work.Vufors 04:44, 13 July 2006 (UTC)
 * In regards to "non-spacial data" see my topic on this page Vufors 04:44, 13 July 2006 (UTC)

[3]. in, for example, a long list, some errors have been found, suggesting that the list as a whole may need further checking.
 * Error free.Vufors 08:52, 12 July 2006 (UTC)
 * PASS the Test. Vufors 08:52, 12 July 2006 (UTC)

[4]. it has been written (or edited) by a user who is known to write inaccurately on the topic.
 * False. The author is a professional, and original author is an authority on this topic.Vufors 08:52, 12 July 2006 (UTC)
 * PASS the Test. Vufors 08:52, 12 July 2006 (UTC)
 * It nearly deserves a, as the primary author of the article is the expert in the field, and has placed references to the article and topic in inappropriate articles. See, for example, Requests for arbitration/Carl Hewitt for an extreme example.  I don't think Friedenr has gone that far &mdash; yet.  As far as I know, we don't have a tag for "Article has been primarily written by the inventor/creator".  I think I'll add Controversial.&mdash; Arthur Rubin |  (talk) 14:56, 12 July 2006 (UTC)
 * Carl Hewitt is a red herring... the two have no connection. Vufors 04:44, 13 July 2006 (UTC)
 * That sound threatening & spitefule.That would be against the Wiki Principals. You can add "Controversial", but you need to argue your case on this page. Vufors 04:44, 13 July 2006 (UTC)


 * Still disputed, I'm afraid, but I won't put the full tag back in without concensus. Arthur Rubin | (talk) 14:56, 12 July 2006 (UTC)
 * That's Ok, your entitled to an opinion. Regards Vufors 04:44, 13 July 2006 (UTC)

Unbalanced Tag - Discussion
As per the Wiki rules (below 3 content policies), please state what part or parts in the said article, are unbalanced (NPOV), so they can be discussed and corrected. Vufors 04:50, 13 July 2006 (UTC)

'''[a]. Wiki Conditions'''
 * - "NPOV (Neutral Point Of View) is a fundamental Wikipedia principle which states that all articles must be written from a neutral point of view, representing views fairly and without bias".


 * - "Wikipedia:Neutral point of view is one of Wikipedia's three content policies. The other two are Verifiability and No original research. Jointly, these policies determine the type and quality of material that is acceptable in the main namespace. Because the three policies are complementary, they should not be interpreted in isolation from one another, and editors should therefore try to familiarize themselves with all three. These three policies are non-negotiable and cannot be superseded by other policies or guidelines, or by editors' consensus. Their policy pages may be edited only to better reflect practical explanation and application of these principles".

'''[b]. Three Content Policies'''

[1]. Fairly and without bias. Per Wiki - "A bias is a prejudice in a general or specific sense, usually in the sense of having a predilection for one particular point of view or ideology. One is said to be biased if one is influenced by one's biases".
 * PASS the Test. - Vufors 07:21, 13 July 2006 (UTC)
 * The Wiki Tests:
 * Readers are left to form their own opinions - FAIRLY = PASS
 * Fairness of tone - FAIRLY = PASS
 * A vital component - good research - YES LISTED = PASS
 * Undue weight - Balanced FAIRLY = PASS
 * Tiny minority should not be represented except in articles devoted to those views - CLEAR = PASS
 * Ethnic or racial bias - WITHOUT = PASS
 * Gender bias - WITHOUT = PASS
 * Corporate bias - WITHOUT = PASS
 * Class bias - WITHOUT = PASS
 * Political bias - WITHOUT = PASS
 * Religious bias - WITHOUT= PASS
 * Sensationalism - WITHOUT= PASS
 * Geographical bias - WITHOUT = PASS
 * Nationalistic - WITHOUT = PASS
 * Vufors 07:04, 13 July 2006 (UTC)



[2]. Verifiability. Per Wiki - "The threshold for inclusion in Wikipedia is verifiability, not truth. "Verifiable" in this context means that any reader must be able to check that material added to Wikipedia has already been published by a reliable source".
 * PASS the Test. - Vufors 06:49, 13 July 2006 (UTC)
 * Read the References.
 * Follow the given sources.
 * Read - Frieden, B. Roy, Science from Fisher Information: A Unification, 2nd Ed. Cambridge University Press, 2004, ISBN 0521009111, pp502 - and follow the working examples & process.
 * CONTACT - You can contact the following Ph. D. physicists/engineers for statements on the validity of the approach: A. Plastino (plastino@sinectis.com.ar ), P.W. Hawkes (hawkes@cemes.fr ), Ray Hawkins (Raymond_Hawkins@Countrywide.Com ) , Stephen Ames (sames@unimelb.edu.au ) , H. Cabezas (Cabezas.Heriberto@epamail.epa.gov ), S. Luo (luosl@mail.amt.ac.cn), R. Venkatesan (r_c_v@vsnl.com )
 * Vufors 06:49, 13 July 2006 (UTC)



[3]. No original research.

Per Wiki - "the only way to demonstrate that you are not doing original research is to cite reliable sources".


 * PASS the Test. - Vufors 06:49, 13 July 2006 (UTC)
 * Univerity of Arizona
 * Cambridge University Press
 * Cancer Research 62, 3675-3684, July 1, 2002
 * Phs. Rev. E 62, 7462-7465 (2000)
 * J. Chem. Phys. 119, 9401-9405 (2003)
 * Plasma Phys. Control. Fusion 38 (1996) 1849-1878
 * Ecological Modeling 174, 25-35 (2004)
 * Fuzzy Sets and Systems 128(3): 285-303 (2002)
 * IJCC, 2 (4): 1-63 (2004)
 * Vufors 06:49, 13 July 2006 (UTC)



Reliable Sources
I would be more impressed if the references for "fundamental" results were written by someone other than Frieden, himself, but I suppose they'll do. As I've mentioned in the case of other articles written by the expert in the field, conference proceedings are not necessarily peer-reviewed, and survey works in "new" fields may only be reviewed by the author, other than for obvious typos. This applies even if the publisher normally produces peer-reviewed works. &mdash; Arthur Rubin | (talk) 05:00, 14 July 2006 (UTC)


 * Hello Arthur,
 * I am going to expand the text out so that Wiki people can get a working knowledge. I have arranged the text, so that it can be expanded. This may take some time, as I have other University commitments. Reviewers are not a good place to get comments from, they only summarise what they see or read, they never confirm it with calculations. It is through application and sources in peer reviewed Science Journals the ideas make it or die. Good luck on your Admin Nomination - Regards to you - Vufors 06:55, 14 July 2006 (UTC)

Material added by B.R. Frieden on Sept. 14, 2006
I am overall happy with the article, except for one needed clarification (response to 2nd sentence of 2nd paragraph). It was a fair remark, and can be squarely answered, as in the material I added today (see below). If you want to shorten it, that's okay, as long as the meaning is not significantly altered. —The preceding unsigned comment was added by 150.135.248.197 (talk • contribs) 20:05, September 14, 2006 (UTC)

The preceding message was posted a few hours ago. Now (5PM, MST, Sept. 14) I see that the material I added to the article is no longer in the article. Was it somehow blocked? Please inform. —The preceding unsigned comment was added by 150.135.248.197 (talk • contribs) 00::05, September 15, 2006 (UTC)


 * The last sentence of the addition is just wrong. "Detection coordinates" (which is a name for a standard concept, so I won't argue much) has nothing to do with the location of the observer.  In fact, in general, the choice of coordinates has little to do with the location of the observer, and conflating those concepts may be a serious error.  There may be something to be said in regard the principle of least action depending on the coordinate system chosen, but it doesn't, in general, do so.  &mdash; Arthur Rubin |  (talk) 13:35, 15 September 2006 (UTC)

B.R. Frieden response, Sept. 20, 2006
Let me address the critical-sounding statement that was added to the article,

"Some critics regard the theory as vague and ill-defined. For example, the location of the extremum may depend on the coordinate system used to model the physical objects."

In response, I wish to make two points:

(1) EPI is not alone in having this apparent (see below) drawback. In general, the ordinary Lagrange action approach suffers it as well (for details, see H. Goldstein, Classical Mechanics, Addison-Wesley, 1950; Sec. 6-6, pp. 207-208). For example, the action approach is only independent of Lorentz coordinate transformation in special cases of the potential. In fact, Goldstein states (on p. 208) that for this reason he will limit his famous development to only two special cases of the potential: the free field and the electromagnetic field (p. 208). The gravitational potential, for example, is ruled out (p. 208).

(2) The Wheeler participancy of the observer that is cited in the article includes the observer's choice of coordinate system. It is by now well known in modern physics that, according to which coordinates are chosen, a different physical effect can be seen. For example, in the optical double slit experiment, if the coordinates so chosen are in the image plane the physics obeys the electromagnetic wave equation, whereas if they are in the slits the physics obeys the Schrodinger wave equation. Thus, the dependence of an EPI output upon the chosen coordinates is a realistic property and not a drawback. We would want a general approach to physics to have this property.

I really would appreciate engaging in a constructive discussion of these matters. A legitimate theory can only profit by it. 150.135.248.197 22:43, 20 September 2006 (UTC)B.R. Frieden


 * As you've edited your response, I cannot comment on point (1), although I believe it to be incorrect. Point (2) is just wrong.  That is a non-standard, and, IMHO, improper interpretation of quantum mechanics, to assert that the choice of "coordinates" effects the results of the optical double slit experiment.  The choice of observation may affect the result, but not the choice of coordinates.  &mdash; Arthur Rubin |  (talk) 18:34, 22 September 2006 (UTC)

(1) Is wrong as well. Quoted right from our own article: "Don't let the use of standard variables such as x fool you--the beauty of the Lagrangian is that you can use any coordinates you want, which don't even need to be orthogonal. These are "generalized coordinates"." The Lagrangian is coordinate invariant. RogueNinja 22:20, 5 March 2007 (UTC)

Citations for the critics
We need some citations from real people and real papers before you can say this. Regards Vufors


 * 1) Frieden's alleged "method" for obtaining Lagrangians appears not to be well-defined . If so, one might suspect that this "method" is in fact merely a tool for constructing ad hoc "hand-waving" derivation of known results.
 * 2) The desired extremum of I − J may not always exist.
 * 3) Frieden appears to be trying to follow the model of Edwin Jaynes, who applied Shannon's notion of information to physics in 1957 (following an even earlier observation of John von Neumann), which led to the principle of maximum entropy. However, while Shannon's entropy has a clear non-parametric rationale, the "information" interpretation of Fisher information is less clear, particularly in the context of Fisher's claims, and apparently limits Frieden to one-parameter models.


 * I also noted this odd statment in the so called ref section. "Note that these are general refs. The criticisms above are not necessarily based on them" - WHAT!!! Not necessarily based... find real stuff folks! How odd? Vufors 04:20, 4 July 2007 (UTC)

After doing some reading today, in regards to this ref the fact that you used "may" tell me that you need to do more fact finding as the two ref are not clear. So I have set it back to the discussion mode. Vufors 08:27, 4 July 2007 (UTC)


 * 1) Shannon's entropies obey several extremely useful identities which not only justify the interpretation of Shannon "information" but play a crucial role in the success of classical information theory. Fisher information does not obey these identities, which may render a theory based upon Fisher information less powerful than one based on Shannon information.

This page should be deleted
Frieden is a total crackpot, and we all know it.81.178.157.195 (talk) 11:47, 31 January 2012 (UTC)


 * I don't know that. His stuff may not be a nice clean theory but I definitely get the feeling there is something there and it is not crackpot. In mathematics the umbral calculus stated off the same way and took over a hundred years before the underlying reason it worked was teased out and it was put on a firm foundation. Dmcq (talk) 14:54, 31 January 2012 (UTC)


 * Who knows if it correct? The explanation in the article is nonsense. 178.38.122.224 (talk) 17:03, 24 February 2015 (UTC)

Hardy's axioms
The claim that "... the mathematical axioms of L. Hardy ... underlie all known physics" requires a reliable source, not one based on this fringe non-mainstream theory. — Arthur Rubin (talk) 22:05, 9 March 2014 (UTC)

Nonsense article
This article doesn't remotely explain its thesis. For example:



\mathcal{I} - \mathcal{J} = \mathrm {Extremum} $$

What are these functions of? What can vary? Instances? Parameters?

Example?

Is there any content here at at all??

Since when does this kind of nonsense get onto Wikipedia? 178.38.122.224 (talk) 16:42, 24 February 2015 (UTC)


 * This might help: http://arxiv.org/abs/1501.00054 . In this paper it is described how the antisymmetric part of the Fisher information for a Kahler structure (on a finite dimensional Hilbert space) has special significance.  It is probably not a great stretch to intuit then that perhaps 'I' and 'J' relate to the symmetric and antisymmetric parts of an appropriate model, respectively, unfortunately as of May 2015 the arxiv seems to be rather thin on examples in this direction geared towards the 'EPI' approach.  Regardless, one can probably then view I - J = extremum in at least this context as just a restatement of the Cramer-Rao inequality, for the special case where the underlying structure of the geometry can be decomposed into symmetric and antisymmetric parts. RogueTeddy (talk) 06:34, 17 May 2015 (UTC)