Talk:A New Kind of Science

From Wikipedia, the free encyclopedia

Contents

[edit] Organization of Criticism, and general reorganization

Hello all. I've recently been making changes to the page. (For those of you who worry about all the links, I will be progressively re-adding them)

I know a lot about NKS and like it quite a bit. But I want the criticism section to be as good and fair as possible. While reorganizing it I've actually added a bunch of criticisms that were not there before. So before flaiming me keep in mind that I want critical viewpoints to be expressed as well (and as reasonbly) as possible.

One thing that troubles me about how I've done things is that the criticism section is all "Critics say X", whereas the response to criticism is written in the first person. My reason for this is because the criticism comes from many different sources, and some of it is self-contradictory. I am just worried that the assymmetry will be percieved as a sign of bias. I am not sure what the proper way to do it is; there are probably precedents elsewhere in wikipedia on how to handle this.

I would also like to have a section with extensive quotes from many different reviews. Perhaps it would be better to break this page up into several pages. For instance, a separate controversy page, and a separate page about the field of NKS rather than just the NKS book proper. -- Kovasb

(I added a signature to make things clearer).
I haven't looked too closely at your recent edits, but I didn't like that you had replaced a nicely-linked text with some text without links (especially since I was the one that came and added the links a few weeks ago ^-^). But hey, since you're still working on it, no problem.
I don't think putting the criticism on another page is standard practice on wikipedia. I tend to find that putting criticism as close as possible to the idea being criticized gives the best results (not that I've extensively tested this) - inserting criticism into sections, into paragraphs, rather than bunching it at the end of the article. I don't think putting too many extensive quotes would be that great (Maybe it could be put on wikiquote, though a link to a place collecting quotes - I think there already is such a link in the article - may be the best. We don't want to bore the readers.
As for a seperate article for the "New Kind of Science" itself, mmm, I'm not sure that terminology is widely accepted (It would probably have been much easier if he had come up with a name like "generative complexity" - something unused). I think a safer step is to integrate material into existing articles, like complexity or randomness or cellular automata or computation ... the book has a lot of relevant, well-researched discussions on those topics. I guess I prefer wikipedia to be a mish-mash of everything rather than having seperate "bubbles" of information weakly linked to each other. Explaining how NKS fits in with systems theory, chaos theory or algorithmic randomness is, in my mind, more important, instructive and interesting than having a complete coverage of the criticism of NKS. At least, that's the direction towards which I've been moving (I'm now mainly working on randomness). Flammifer 04:46, 19 September 2005 (UTC)
I think the idea of exporting material into other topics is a fairly sound one. I'll try to work on that. "Generative complexity" is an interesting phrase. I've thought of "pattern generation" as a potential subfield, but it doesn't quite capture the entire set of ideas.

[edit] Von Neumann CA

Last year I saw that someone had developed a CA which was capable of adding two numbers together. (which is a von neuman computer). I wish I had taken more notice: but, I was burning up the cpu with CA and could be bothered to look at it's static structure. There were 2000+ parts and like I say it was static. And I'm all about the visuals.

This, in my eyes, lends support to his thesis

Most important thing I learned was that western science since Newton and differiential calculus, has had a bias towards equations. --Two16 18:42 Jan 15, 2003 (UTC)

Conway's Game of Life has long since been able to add two numbers together, as it is known that a Turing machine is reducible to it. This doesn't really support his thesis (that a Turing machine is reducible to any "sufficiently" complex system), except in that it doesn't contradict it...
(As for von neuman, I'm not sure if you're talking about the Von Neumann architecture, i.e. modern processors, or a von Neuman machine, i.e. self-replicators. This is neither.)
--Andrew 68.6.190.89 08:36, 14 April 2006 (UTC)

[edit] Pseudoscience

I have removed this:

  • "It is for these failings and for Wolfram's penchant for excessively grandiose claims that some have voiced that they would consider this work to be a piece of pseudoscience if it had a wider following."

The book is already heavily attacked, and to call it pseudoscience is not correct. It might be called incorrect or flawed but it doesn't fall under the usual concept of pseudoscience. Wolfram has a Phd in Physics and he has written peer reviewed articles on cellular automata. If you want to put anything like this back in the article please post solid reason here.

I was honestly trying to find a more moderately phrased version that would still contain vital and unassailably NPOV information. The original text was not mine, and definitely too pointedly phrased. I don't however accept that Wolfram should get a "bye" just because some have already (justly) confronted his ideas and way of presenting them elsewhere.
Wolfram's other writings may have been scientific, but that does not make A New Kind of Science scientific, any more than Isaac Newton's Principia Mathematica being science makes his Chronologies (a prophetic work derived from silly "analysis" of the bible) scientific. I have nothing against Wolfram's tome, and think in the final analysis it will fare better against its critics than you seem to have confidence in. But that is not a reason to not characterize its current reception accurately. I won't get into an edit war over this. But this is how I feel --Cimon Avaro on a pogo-stick 18:21 21 Jul 2003 (UTC)
Well, it's repelling because it dares to posit a music of existence. Letting aside for a moment its "validity", the fact that it dares to do such a thing is sufficient to attract repulsion. The truth it posits cannot be empirically "tested" because the truth it's describing comprise the toolkits used to test that truth. The grammar of cellular automata subsumes itself. --Gyan 09:54, 18 Apr 2004 (UTC)-


[edit] POV formatting

  • Notable for its publicity and the number of negative reviews in "reputable" journals,

Why is "reputable" in quotes and italics?? The AMS Notices (I believe this journal published a critical review) is a reputable journal, not a "reputable" journal. Regardless of whether you agree with the review. And they weren't the only ones. The quotes and italics obviously are meant to imply that the only journals that would publish a negative review would be journals that aren't reputable. --Revolver 03:53, 29 Feb 2004 (UTC)

The original formulation stated "uninfomed reviews in 'reputable' journals". I believe its because book reviews are themselves not peer-reviewed. In particular, book reviews complaining that NKS is not peer-reviewed were not peer-reviewed, thus resulting in a number of extremely basic conceptual and even factual errors. Thereby besmirching the reputation of the their publisher.


[edit] Criticism of basic premise

Lumidek replaced "The second type of criticism comes from people who cannot even accept the book's basic premise", with ""The second type of criticism comes from people - for example those who understand quantum physics - who do not accept the book's basic premise," citing "arrogant formulation "cannot even accept" changed to "do not accept". Surely, Lumidek realizes that his/her phrasing is itself arrogant since it implies that people who (accurately) "understand" quantum physics cannot accept Wolfram's premise. --Gyan 07:41, 3 Jun 2004 (UTC)-

Well, Gyan, you can describe this statement in any way you want (as arrogant?) and you may try to act as a censor. Nevertheless, these are the only things that you can do against the important fact that the people who accurately understand quantum physics cannot accept Wolfram's premise simply because the local hidden variables theories - which is a class that easily contains every single model that Wolfram has ever thought of - have been ruled out decades ago, via Bell's theorem. OK, I will try to create a new formulation that could satisfy you. Please try first to show us a professional quantum physicist who will declare that the world may be based on these rules, send me his e-mail to motl@feynman.harvard.edu and consider back-editing only afterwards. See also EPR_paradox#Modern_perspectives_on_the_EPR_paradox. All the best, Lubos Motl, 8.24 (Eastern time)
Wolfram's proposed networks do not, in fact, require purely local variables. Bell's theorem, per our own page on it, does not apply (Or more accurately, does not disagree) if information can travel faster than light. The networks Wolfram suggests would restrict MOST information to travelling "slower than light", but it's perfectly possible for there to be some sort of entanglement - his networks, after all, do not explicitly represent space, and the emergent property that he claims produces space would, in fact, not be a barrier to entanglement. Michael Ralston 06:40, 20 February 2006 (UTC)
Well, my QP understanding is a bit rudimentary, but does this(PDF) help? --Gyan 17:32, 3 Jun 2004 (UTC)-
Dear Gyan, I have already won a couple of bets based on my statement that people will continue to agree that these 't Hooft's papers are wrong and uninteresting. Most great physicists get a little bit insane and write crazy papers like that when they become older - you were just unlucky to choose an example. 't Hooft's recent papers don't change a single bit about Bell's inequalities and other statements made above. If you want a layman's tool to decide whether you should consider a paper seriously, it is called "citations". On http://www.arxiv.org/cits/quant-ph/0212095 click "cited by" and you will see about 5 citations (from suspicious papers). Serious and important papers have abruptly hundreds of citations. If you wish, I can show you hundreds of papers that are 50x more important than the 't Hooft's paper you've chosen, including mine. --Lumidek
"The second type of criticism comes from people who cannot even accept the book's basic premise" sounds as if the article is taking Wolfram's side (POV). Lumidek's 2nd wording looks best to me, more detailed and neutral. --Andris 17:56, Jun 3, 2004 (UTC)
Two points. First, for those who have not read or understood Chapter 9 of NKS: Bells Theorem does not apply to Wolfram's model. Why? Because the nature of space is unlike that for which the theorem was formulated. In fact it is fair to say that in Wolfram's model space is not even directly represented - it is an emergent consequence that is observable to those within the network. (I am also interested to know how Lumidek knows about every model Wolfram has ever thought of, and instantly knows everything about them: that they must satisfy his/her assumptions)
We should distinguish two very different situations. One hypothetical situation is that Bell's inequalities do not apply to Wolfram's models; and the other, very different situation is when Wolfram writes a sentence in his book that they do not apply. Of course that they do apply; they were constructed exactly for this kind of simple models. This sort of "cellular automata" (although in a continuous edition) is exactly what de Broglie, Bohm, Einstein, Bell and others were thinking about when they were proposing hidden variables, and the proof of Bell's theorem - and the experimental confirmation of their violation - just killed this class of theories that includes Wolfram's (in fact, Wolfram's is a simplified version of the old hidden variable models because it does not even use the real numbers). Moreover Bell's theorem does not care how space is exactly represented; it cares whether the theory is local. Wolfram's models have their notion of locality; either this locality agrees with the physics definition of locality in space, and then Bell's theorems apply, or they don't, but in the latter case Wolfram has no idea how the apparently local or quasi-local physics in space could emerge. By the way, I am probably closer to knowing Wolfram and his motivations etc. than you think. For example, Andrew Strominger is his good friend, and you can investigate the relation between Andy and me. All the best, Lumidek
This is not a cellular automaton model. Nor is it like the various hidden variables models that have been proposed. There is no distinction between space and matter, and time does not update uniformly. This is a pretty serious departure from the common notions of how the universe works. And it is certainly not the kind of model Bell had in mind when designing his work. The way that these theorems work depends on matter being in definite locations in space. But the point i keep trying to make is that the notion of space is fairly ambiguous in Wolframs model, and is not directly represented. A consequence of that it is that is trivially possible for information to instantly travel what appear to be arbitrary distances. In passing you seem to conceed that Wolfram's notion of locality may not agree with THE physics definition. Wolfram does have an idea about how quasi-local physics can emerge, although obviously not a complete model. And whether or not his ideas turn out to be correct or incorrect in the long run is irrelevant to the discussion of whether or not they are immediately disqualified in the most obvious and embarassing way by a fundamental quantum mechanical issue that every undergrad knows about. kpb
Second point: Fundamental physics as a simple program is an application of NKS that takes up half a chapter out of 12. It is wrong to say that a fundamental premise of NKS is that the world "is" a simple program. Yet again people are confusing themselves about what modelling is. In the case of the fundamental theory of physics, Wolfram believes that the universe literally may be a simple program. But in all other cases, simple programs are just a tool to model what nature is doing, much as we use equations to describe what nature is doing. Therefore any conclusions about whether Wolfram's specific ideas about fundamental physics turn out to be true are irrelevant to the ability to model nature with simple programs in useful way. The idea that hidden variables arguments somehow impact one's ability to for example make the model of leaf growth in chapter 8 is rediculous. --kpb
Just for the record, Wolfram directly addresses the Bell inequalities in NKS, starting on p. 1064. The criticism makes it look like he ignores the subject. --Wfaxon 10:38, 8 December 2006 (UTC)

[edit] 't Hooft

  • Dear Gyan, I have already won a couple of bets based on my statement that people will continue to agree that these 't Hooft's papers are wrong and uninteresting. Most great physicists get a little bit insane and write crazy papers like that when they become older - you were just unlucky to choose an example. 't Hooft's recent papers don't change a single bit about Bell's inequalities and other statements made above. If you want a layman's tool to decide whether you should consider a paper seriously, it is called "citations". On http://www.arxiv.org/cits/quant-ph/0212095 click "cited by" and you will see about 5 citations (from suspicious papers). Serious and important papers have abruptly hundreds of citations. If you wish, I can show you hundreds of papers that are 50x more important than the 't Hooft's paper you've chosen, including mine. Lumidek

The number of citations is not a measure of the correctness of the claims made in the paper. It isn't even a measure of how interesting the paper is. To cite 't Hooft's paper you must write a paper on a deterministic model underlying Quantum Mechanics (and also use 't Hooft ideas). Since his theory is so completely different from the theories most physicists are working on, it requires a major effort for an interested physicist to produce a paper citing his work.


As far as I know no papers have appeared proving 't Hooft's theory wrong. In fact, as 't Hooft correctly argues, there are several loopholes in Bells theorem. 't Hooft favors pre-determinism. Bell's theorem isn't applicable in a completely deterministic setting (Bell himself made that point too). What the observer decides to measure is already fixed in a deterministic model. So, one cannot say that the observer could have decided to measure something different.

Pre-determinism is a rather awkward concept, and I have to admit that I don't like it. It is however, trivial to give an existence proof of a deterministic theory underlying Quantum Mechanics. Consider a large classical computer simulating our quantum world to some reasonable degree of accuracy. Then the true laws of physics describe how the computer works and these are deterministic laws. They are local at this level (i.e. the switches of the computer interact with each other in a local way).

Be sure that 't Hooft's papers that are usable and appreciated by the scientific community are cited. The missing citations are not, of course, the reason why I think that the papers are wrong. They are just my recommended tool to decide for all those who are offered a lot of material and they don't have enough knowledge to decide according to the content. Moreover, the physics community is a peaceful one and 't Hooft has done a lot for physics. I see no point of writing a "paper" why 't Hooft is wrong, because everyone knows it, and no one wants to put 't Hooft into a bad situation because we like him. Be sure that 50 percent of my colleagues would be immediately able to learn one or two missing details and write papers about 't Hooft's proposals if they found them to be attractive in some way. These papers are pretty trivial - look like popular talks - and you should not really believe that they are too complicated for other physicists. Finally: Saying that Bell's theorem does not apply to deterministic models is like saying that the U.S. constitution does not apply in America. Bell's inequalities are theorems exactly ABOUT the local deterministic (or at least "realistic") models. --Lumidek 15:19, 25 Oct 2004 (UTC)


  • Saying that Bell's theorem does not apply to deterministic models is like saying that the U.S. constitution does not apply in America. Bell's inequalities are theorems exactly ABOUT the local deterministic (or at least "realistic") models
In the derivation of the Bell inequalities it is crucial to assume that the observer could have decided to measure a different component of the spin, while leaving everything else the same. Strictly speaking this cannot be the case (pre-determinism see above). Most physicists believe that replacing the real (deterministic) observer by one that can magically violate the laws of physics is allowed.
However, this issue hasn't been resolved. No one has been able to prove that pre-determinism isn't a loophole. This is not a crazy concept invented by 't Hooft, but was invented by Bell himself (he gave a list of all possible loopholes in one of his papers).
  • I see no point of writing a "paper" why 't Hooft is wrong, because everyone knows it, and no one wants to put 't Hooft into a bad situation because we like him. Be sure that 50 percent of my colleagues would be immediately able to learn one or two missing details and write papers about 't Hooft's proposals if they found them to be attractive in some way. These papers are pretty trivial - look like popular talks - and you should not really believe that they are too complicated for other physicists.
A few of his papers were actually talks at conferences. They are certainly not difficult to understand. His ideas are interesting to the 'quant-ph' people who work on issues related to interpretation on quantum mechanics. I note that they haven't been able to prove 't Hooft wrong. Predeterminism as a loophole is criticised here on pages 19-20.

[edit] Strings from Logic?

Strings from Logic


[edit] 81 citations!

'T Hooft's first paper on deterministic theories has 81 citations!

[edit] article too long, and smacks of boosterism

I just did some copyediting on the first half of the article, specifically removing 'NKS' abbrev, and minor rewriting where needed. More needs to be done.

May I say the article was clearly written by a proponent. I think it could stand some shortening — why does wikipedia need such a long treatment of this book?

Criticism should be beefed up more, and the responses to criticism I think should be interleaved with the criticism itself.

Let me propose a criticism: why does science need a new branch to study these problems? They seem to fit quite well into cybernetics (systems theory).

--

I agree it is too long. Some of the material should be exported to other areas.

It was written by a proponent (me).

Concerning this new potential criticism, it would fall under the category "Originality and Self-Image". There are many such fields.. why couldn't it be math, why can't it be physics, why can't it be computer science, etc.

--

I also think the article is too long and doesn't seem balanced in the detail it contains. The treatment of the core ideas from the book does not need to be exhaustive - that's what the book is for, not wikipedia. The "response to criticism" section seems unnecessary for a balanced overview of the book. I suggest it be removed and really important bits can be incorporated into the criticisms section (but let the criticisms stand!). Once the article starts to read like a debate, something must be wrong...

[edit] Criticism / response

Hello. I'm going to cut the "Response" stuff from the criticism section. WP articles on controversial subjects aren't generally structured like that; it's not WP's business to defend Stephen Wolfram, or anyone else, from his critics. 64.48.158.87 04:09, 17 April 2006 (UTC)

Sound good. Go ahead. (And why not get an account—that would make communication much easier.) Best, Arbor 09:50, 17 April 2006 (UTC)

[edit] The introduction

The introduction should be clearer. What does Wolfram want to tell us with his book, what is his message, his main idea, what does he wants to show? The article is supposed to be read not only by those who are familiar with the subject, but by everybody who is interested to see what it is all about. If I undersand it correct, he claims that the universe is digital and based on information, and scientists should not try to describe the nature through mathematical formulas and equations, as they have done so far, but through simple cellular automata (probably best described as/compared with computer programs). This goes not only for physics, but also for chemistry and biology; all forms of science who describes nature should have this single common platform to branch out from. —The preceding unsigned comment was added by 193.217.193.228 (talk • contribs) 21:37, 9 February 2007 (UTC)

[edit] Kudos to those who have contributed to this article

Amidst the fog of stimulating but uniformative "Wolfram is a genius!" "Wolfram is a nut!" debate, it's a pleasure to see an article that states, concisely but clearly, what Wolfram is actually saying. -- Writtenonsand 18:48, 4 May 2007 (UTC)

[edit] If I have understood it correctly

What Wolfram means is that the universe and everything in it is digital in its nature, and can (and should) be described as programs rathers than mathematical tables and formulas? If so, it should be explained clearer in the introduction since it really is a revolutionary way of observing the reality, correct or not. 217.68.114.116 (talk) 17:35, 17 December 2007 (UTC)

[edit] /.

For those as understand these things, the links provided at this /. discussion might be of use. .. dave souza, talk 16:30, 1 January 2008 (UTC)

[edit] Fair use rationale for Image:Newsciendfsdfs.jpg

Image:Newsciendfsdfs.jpg is being used on this article. I notice the image page specifies that the image is being used under fair use but there is no explanation or rationale as to why its use in this Wikipedia article constitutes fair use. In addition to the boilerplate fair use template, you must also write out on the image description page a specific explanation or rationale for why using this image in each article is consistent with fair use.

Please go to the image description page and edit it to include a fair use rationale. Using one of the templates at Wikipedia:Fair use rationale guideline is an easy way to insure that your image is in compliance with Wikipedia policy, but remember that you must complete the template. Do not simply insert a blank template on an image page.

If there is other fair use media, consider checking that you have specified the fair use rationale on the other images used on this page. Note that any fair use images lacking such an explanation can be deleted one week after being tagged, as described on criteria for speedy deletion. If you have any questions please ask them at the Media copyright questions page. Thank you.

BetacommandBot (talk) 22:48, 2 January 2008 (UTC)