Talk:Butterfly effect/Archive 1

Archive 1

Archive 2

Archive 3

The issue of disambigaution

I wonder if we need a disambiguation page? --Jeff 06:20, 30 May 2004 (UTC)

How can the error bound exceed 100%? Wouldn't that mean it's wrong more than all of the time? — Daniel 18:44, 15 Feb 2005 (UTC)

The "action drift over time" section was not encyclopedic, so I removed it. I tried to clear up some confusing parts, but it could still use some work. Should there be at least some mention of the cultural references present in some of the earlier edits? Comments/suggestions welcome! — BryanD 03:46, Apr 26, 2005 (UTC)

Is it not possible to have a more than 100% error? In chemistry, if you are supposed to have yielded say 10 grams of a certain chemical, and you yield 25, is that not a percent error of 150?

yes, it isn't...Banno 08:38, Jun 25, 2005 (UTC)

Stock market, Etc.

I've removed stock market as an example: problems in prediction of that have more to do with knowledge affecting the result than with butterflies. Also revised the why NWP is difficult bit. Again, butterfly level stuff isn't really relevant in practice. William M. Connolley 10:26, 10 October 2005 (UTC).

Just as much as "lucy cakes" and "the horoscope". Drivel! But I believe in domino's, however, but tryin' to potray Butterfly's as the "domino effect" of Nature, y'right. Perhaps on a more global level, I guess. Big winds and/or tidal waves, sub-surface drifts, etc. But flappin' your wings...only as gettin' a fan in'yo face. Simple as THAT! It's no use in speculasation'--OleMurder 23:58, 12 December 2005 (UTC)

Sound of Thunder

Why is concept described as "abused" by popular media dealing with time travel? The time travel itself is irrelevant, it only provides a way of changing the initial conditions. The butterfly effect is a consequence of unpredictability over a sufficiently long time scale, it doesn't matter what the start and end times are, nor what model is used, etc.   — Lee J Haywood 22:40, 19 November 2005 (UTC)

Its abused because typically (as in "sound of thunder") they go back in time and are really really careful not to crush any butterflies. But! Just their mere presence (let alone the floating walkway) would be enough to totally change the weather and hence all history. William M. Connolley 01:01, 20 November 2005 (UTC).

Ah, A Sound of Thunder was already on my list of things to watch but I didn't know from the article that the characters were trying not to change anything. I have seen The Butterfly Effect which, to be fair, mainly involves a character that is deliberately trying to change things – even if the results are unpredictable. I've used some of your words to clarify the article a bit. Thanks.   — Lee J Haywood 09:44, 20 November 2005 (UTC)

I disagree with simply changing "used" to "abused" without further explanation, as this implies that these uses of the term in science fiction constitute "abuse" (i.e. a misuse or fundamental misunderstanding of the term). It implies that these things are inappropriate uses of the term butterfly effect. As I understand it, these examples have the basic notion correct ... that seemingly insignificant changes can produce long-term effects that are both huge and unpredictable. The fact that the details are not completely accurate (underestimating just how insignificant the changes can be and still produce enormous effects) falls within the normal parameters of science fiction. I recommend (1) changing "abused" back to "used", but (2) adding a paragraph about why the use is often inaccurate in its details.   — F Bunting 19:05, 22 November 2005 (UTC)

Second point (for my own clarification), Lee wrote: "Just their mere presence (let alone the floating walkway) would be enough to totally change the weather and hence all history." Changing "would" to "could" is more accurate. True?   — F Bunting 19:05, 22 November 2005 (UTC)

I agree – though it was William that mentioned the walkway, not me (my fault for not indenting properly, I suppose). I've changed the article to take account of your comments, though I'll leave it to someone else to explain why popular use is often inaccurate. Thanks.   — Lee J Haywood 23:20, 22 November 2005 (UTC)

No, it was my fault for not reading properly. Thanks for making the change. This is my first foray into wikipedia and didn't feel like editing without discussion (is that the norm?). Incidentally, I also thought it unfair to let Sound of Thunder stand as the prime example of the "abuse" of the theory, seeing as how it is based on a story written 10 years before Lorenz worked out the details of the theory. I think it's fine now.   — Frederick B 23:37, 22 November 2005 (UTC)

If you don't vandalise and try to keep typos to a minimum, you can change any article any way you like. No-one owns any article and if anyone who happens along doesn't agree with something they can, and will, simply change it. Even if you mess up and totally remove or corrupt the content of an article, it will simply be reverted to the previous version.

On a different note, I was wondering if it would be reasonable to mention Sliding Doors in this article? Though technically there is a massive change right at the beginning (missing the train in one reality, catching it in the other), the butterfly effect starts a few moments earlier when she is slowed-down fractionally by a boy on the stairs – but only in one reality. The other point is that neither reality converges, and there are the effects that she has on other people, that propagate to yet more people. Thanks.   — Lee J Haywood 19:38, 23 November 2005 (UTC)

Rv: Why

I reverted this. The anon version is garbled. NS isn't a good source anyway. Certainly, initial condition error is at a vastly greater scale than butterflys wings. There *is* some stuff about the relative sizes of model and initial error. But saying that BE is a figleaf is wrong.--William M. Connolley 18:28, 26 November 2005 (UTC).

Utterly wrong

Just like everyone else, we got it utterly wrong. Although Lorenz analyzed the butterfly effect in the '63 paper, he didn't give it any fancy interpretation (and certainly not its name) in terms of butterflies and tornadoes there. I checked it myself. See this. Or, if that doesn t work: Journal of Neuroendocrinology Volume 16 Issue 1 Page 1 - January 2004

Of course the now popularized fanciful notion of the butterfly effect can't possibly work. Most people intuitively understand this. From a practical standpoint if it were true, weather would degenerate into a completely chaotic, extreme system with no calm periods and be nothing less than a continuous series of tornadoes one after another in all parts of the globe. The concept ignores the fact of frictional losses in the air causing the air motion to be damped out. The miniscule air disturbance from the butterfly’s wings dies out and causes no effect whatsoever. But this is the kind of romanticized pseudoscience that makes more common-sense oriented people shake their heads and wonder what Scientists do all day. D. Whyte

You are wrong. The butterfly effect works as advertised, though its fairly easy to misunderstand it and most people do. See [1] and [2] (yes I wrote them). William M. Connolley 16:24, 23 December 2005 (UTC).

I suspect D.Whyte's confusion is between cascading effects and sensitive dependency on initial conditions. I.e. an implicit mental model in which if small things can have large effects, large effects must have still larger effects. I would have thought that the phase space graph I added to the page would help the conception.

I wonder if it would be worth pointing out a really simple example/analogy. One that comes to mind: A ball placed near the top of a hill might roll into any of several valleys on different sides (attractors). Right at the crest, which direction it rolls will be affected by very small difference in placement, or even by other small factors like a micro-current of air (or a butterfly flapping nearby). Lulu of the Lotus-Eaters 18:52, 23 December 2005 (UTC)

A while back I added the simple exampe of a ball sitting at the crest of a hill. In a complete brain glitch, I described that as "non-dynamical" when I meant to write "non-chaotic". An anon editor recently took out the adjective altogether, and just made the example and example. It's probably good as it is, but I wonder if other editors think the first paragraph should alreay try to distinguish strange and non-strange attractors... or is that way too much baggage? Lulu of the Lotus-Eaters 19:10, 1 March 2006 (UTC)

I agree; I've just re-edited to make the point that simple and easy-to-understand systems can have this property William M. Connolley 19:45, 1 March 2006 (UTC)

I reinstated the example of the butterfly beating his wings in the introduction. It's certainly not quite correct as far as the theory goes, but since it is the most popular etymology of the term I thought it deserved mention. --TheOtherStephan 05:24, 8 April 2006 (UTC)

I took it out. We already present the origin a paragraph later, and using the actual example locations from Lorenz' paper, and without all the spelling errors (it's funny how everyone seems to remember different geographic places for the illustration; the concept is the same regardless, but we might as well cite the actual originating one). Lulu of the Lotus-Eaters 06:14, 8 April 2006 (UTC)

I thought I remembered reading in James Glick's book that the term Butterfly effect actually comes from the pattern made by graphing the trajectories, which resemble butterfly wings (as shown in the article). Did I imagine this? Can anyone confirm or disprove?

Poe

Look at the story The Power of Words by Edgar Allan Poe for a fictional reference. Am I just imagining this or did Poe really talk about the butterfly effect in 1850?--85.100.240.148 21:35, 21 December 2005 (UTC)

Didn't see a butterfly in there? William M. Connolley 16:24, 23 December 2005 (UTC).

Not mentioning the butterfly, yes, but the point is the same, right? It's the words instead of the butterfly's wings here..--85.97.87.224 23:54, 2 February 2006 (UTC)

"Before the technologist walks the artist, before the artist walks the child."

I love Poe, he was a genius. But the Butterfly Effect has arguably been understood for centuries, if not longer, though of course it was not called by that name. Remember George Herbert (1593-1633): For want of a nail a shoe was lost, for want of a shoe a horse was lost, and so on to the loss of the kingdom. What is that but a succinct expression of the Butterfly Effect in an intuitive fashion? 192.139.140.243 04:26, 10 February 2007 (UTC)

Animation

User Tó_campos has added a series of images of the Lorenz attractor to the Wikipedia. I like the images, but I find the GIF animation difficult to follow. Is that my browser? XaosBits 02:56, 10 March 2006 (UTC)

I also find the animation to be odd. Rather than a path animation of the sort I've seen before, it just kinda flashes between different parts of it (time segments?). Overall, I do not think the animation is helpful, at least not what I see (or apparently XaosBits too)... perhaps some particular browser version does something different... but I'm going to take it out pending discussion on this talk page. Lulu of the Lotus-Eaters 04:10, 10 March 2006 (UTC)

In fact I made the animated gif file just by using print screen in 3 images of the animation of my JAVA applet in http://to-campos.planetaclix.pt/fractal/lorenz_eng.html. Maybe I can try to do an animation with further images later to make the anomation more interesting. Tó_campos 11:51, 10 March 2006 (UTC)

I think you would find that just the 3 images side-by-side would be more helpful William M. Connolley 12:21, 10 March 2006 (UTC)

Thanks for the change, Tó_campos. I think the three separate images are much easier to understand. It's certainly nice to illustrate the Lorenz attractor, I just found the 3-step animation difficult to follow. I might tweak the arrangement slightly... or maybe create a custom down scaled version of them to get good anti-aliasing. Lulu of the Lotus-Eaters 18:37, 10 March 2006 (UTC)

Why is that image even here? What does the Lorenz attractor have to do with the butterfly effect itself? Just because the attractor coincidentally happens to look something like a butterfly doesn't mean it's related. Nibios (talk) 19:37, 15 July 2008 (UTC)

Recurrence link

Why does the link pictured as "recurrence" point to http://wiki.alquds.edu/?query=Ergodic_theory, an even more technical article? I understand vaguely that the mathematics might explain recurrence, but the link was pretty confusing for me, a casual reader. Icewolf34 14:53, 14 July 2006 (UTC)

I changed it to Poincaré recurrence theorem, which has an explanation of recurrence in dynamical systems. Recurrence is often discussed in the context of invariant measures of the phase space of a dynamical systems, the topic or ergodic theory. XaosBits 04:54, 15 July 2006 (UTC)

controversy

I recall reading soemthing about the Butterfly Effect meaning everything was connected. Much about hte time of the Conference in RIO when sustainabilit development came to the fore. there was a debate with soem saying it was limite dot a very smal lsituation (localized) and others saying any effecet wasmuch wider (i.e. Global) there are aspect s here like climate change, nuclear isotopes in eth atmosphere etc. However this is a controversy ewr shoudl mention.

I think it has the technical (religious / philosophical as well as physics term) name of NON-LOCALITY

http://wiki.alquds.edu/?query=Nonlocality

perhaps soemeon coudl clean up - i just add the reference here. markus petz

Sorry - nonlocality has nothing (strictly) to do with the Butterfly effect. The butterfly effect boils down to "initial conditions matter". Nonlocality deals with (faster than light; hence nonlocal) interactions between quantum objects. I think you may be confusing the genuine randomness (so far as we can tell) of quantum mechanics with the apparent randomness of the butterfly effect. The latter is actually completely deterministic and predictable (so long as we know the initial conditions perfectly - which we never do). This is a common mistake as the two are often discussed together. Cheers, --Plumbago 13:44, 28 July 2006 (UTC)

Mathematical definition

I can't say I understand this, I guess it's because I'm lacking a little context. What does it mean to say that points x and y are both from neighbourhood N? They start off close together initially, I guess, but surely there's a definition of how close...? JulesH 14:46, 1 October 2006 (UTC)

Tis a maths concept. I've wiki'd it so you can check up if you like William M. Connolley 15:27, 1 October 2006 (UTC)

Possible contradiction between this article and Chaos Theory article

In the Chaos Theory article, section 1 Chaotic dynamics it is mentioned: Sensitivity to initial conditions is popularly known as the "butterfly effect", so called because of the title of a paper given by Edward Lorenz in 1972 to the American Association for the Advancement of Science in Washington, D.C. entitled Predictability: Does the Flap of a Butterfly’s Wings in Brazil set off a Tornado in Texas? The flapping wing represents a small change in the initial condition of the system, which causes a chain of events leading to large-scale phenomena. Had the butterfly not flapped its wings, the trajectory of the system might have been vastly different. (bold is my)

The facts in this paper in the History section mentioned: term "butterfly effect" itself is related to the work of Edward Lorenz, who in a 1963 paper for the New York Academy of Sciences noted that "One meteorologist remarked that if the theory were correct, one flap of a seagull's wings could change the course of weather forever." Later speeches and papers by Lorenz used the more poetic butterfly (bold is my)

Maybe the two facts are right but I think someone should clarify this. —The preceding unsigned comment was added by 189.161.16.60 (talk) 03:37, 23 January 2007 (UTC).

Jurassic Park

Do we really need a cite for that one? It's explicitly stated in the film, so why not just watch it if you don't believe the article? That's a primary source. RobbieG 10:49, 9 April 2007 (UTC)

Movies, etc.

Shouldn't "Butterfly Effect" and "JUrassic Park" be listed under the movie subheading? Especially since Butterfly Effect is a movie, and not a TV show Robhakari 04:22, 25 May 2007 (UTC)

Could it actually work?

I think the article could do with some discussion of whether a butterfly flapping its wings really could significantly affect the weather. This question surely has some importance on its own outside the general discussion of chaotic behaviour in the article because it seems to be commonly believed that it could happen, although I've never seen this justified anywhere. Have any proper studies been done to test the conjecture?

Pagw 09:30, 30 June 2007 (UTC)

adding Back to the Future to the movie section

i wanted to suggest adding Back to the Future to the movies section, though dialogue in the movie can be less than technically acurate at times, the movie does have Marty go back in time and change events that alter his "present day". much like a very cheese, but well loved butterfly effect. [[]] 15:52, 24 September 2007 (UTC)

Back to the Future does not show obvious butterfly effects because the cause and effect changes are always obvious. There's nothing subtle about any of them. Wryspy 20:58, 24 September 2007 (UTC)

Adding a Different Movie

Not sure how this works but just thought the movie "Sliding Doors" (G. Paltrow) illustrated the "Butterfly effect" showing how her choices affected the outcome of her life......she was going to leave the bastard boyfriend anyway though........ 124.189.212.7 04:22, 11 November 2007 (UTC)

Sliding Doors has been removed from the page. It does not mention the butterfly effect and does not illustrate indirect subtle changes. Every change involves visible cause-and-effect chains of events. Wryspy 22:45, 15 November 2007 (UTC)

Removed two sources

It may seem paradoxical that I've removed two sources, but I'll explain.

Firstly I've removed a reference, and its supporting source, to the discontinuity in which it snows in one scene but doesn't snow in its equivalent scene in the other timeline. This is in the 1946 movie It's a Wonderful Life. As such, reference to the butterfly effect is somewhat anachronistic and, although it's just barely plausible, we'd need an exceptionally strong source to support the idea that Capra or his writers intended to show that the weather itself was influenced by George Bailey's absence. What we got was a chap's opinion on a website. I've removed the reference to that scene and the source, for that reason.

MY NAME IS CLIFFY TUCKER —Preceding unsigned comment added by 66.142.212.233 (talk) 21:58, 12 March 2008 (UTC)

There is also a reference to a plot summary of The Butterfly Effect at a website that solicits such summaries from its users. The summary was written by one such user. As it happens we have an article on the movie on Wikipedia, edited by multiple people. The movie spoiler site obviously isn't a reliable source, and the plot summary there is not more authoritative than the summary on our own article, so I've removed it. --Tony Sidaway 13:35, 27 October 2007 (UTC)

Sources support that the scene exists and show that someone has made the argument. Feel welcome to find a better source, but there's no need to leave it unsourced.

The snowfall changing is the one example that overwhelming fits the butterfly effect, whether the filmmakers intended it or not. Everything else that's different between those two versions of that world can be attributed to obvious cause-and-effect. The snowfall example also has the beauty of fitting because it, like the alleged butterfly, invokes a possible influence on the weather.

Anachronistic nature of the term is irrelevant. Gravity existed before Newton had a theory about it. Endorphins existed before anyone gave them a name. No one is saying the filmmakers were invoking the term itself -- and it does say arguably, after all. Wryspy 21:11, 10 November 2007 (UTC)

WikiProject class rating

This article was automatically assessed because at least one WikiProject had rated the article as start, and the rating on other projects was brought up to start class. BetacommandBot 09:45, 10 November 2007 (UTC)

"cause"

Someone added, and I cut:

Thus, given two otherwise identical systems, small differences in initial conditions can give rise to later differences between them that are huge. This is not the same as (falsely) proposing that those small differences are solely the cause of the later differences in either one of those systems. In popular imagination it is often put that the butterfly is the cause of a hurricane. This is of course not the case; the butterfly being a part of the system which as a whole will give rise to the later conditions. We can say that if (somehow) the universe had been different such that the butterfly had not flapped it's wings as opposed to had - then the universe at a later stage would have been very different. But this is true of any other such difference we care to propose, including inanimate happenings. We cannot blame or regard as a cause of the later system as a whole the non-happening of that imagined alternative.

I think this goes too far in asserting that the butterfly *cannot* be regarded as a cause.

Consider for a moment a simpler system - a pencil balanced on its point, in a vacuum. One little molecule hitting the tip could be enough to tip it over. Of course, it wouldn't fall without more powerful forces also occurring, but considering the molecule the "cause" is reasonable.

Now consider the real weather, and a hurricane has just occurred. Go back in time 1 month, add the said butterfly, and there would be no hurricane. In that thought experiment, it seems reasonable to call the butterfly the "cause", even though countless other things could also be considered to be the "cause" too. The cut para is trying to say something sensible, but I think it goes wrong William M. Connolley (talk) 21:09, 6 January 2008 (UTC)

The article needs to draw the distinction between differing initial conditions leading to differing later conditions and b) the huge leap beyond that that says that that part of the initial conditions which differs from the alternative be held up as the cause of the later system. Rewrite it - but it should be there.

Butterflies causing hurricanes is of course nonsense:

If a universe happened to pop into existence consisting solely of a pencil in a vaccuum with a particle heading toward it and moreover which had no history before this and which ceased to exist after the toppling over - you would be considering the whole of the earlier conditions and the whole of the later conditions and saying one is responsible for the other.

To repeat the point:

(1) The later state is the product of the WHOLE of the earler state - not some small PART of the earlier state you declare important - a butterfly - or say a particular cubic inch of the atmosphere you arbitrarily imagine could have been in a different state.

KBuck

I disagree with most of that. To take one point, your last assertion clearly goes too far - it does not allow us to say that pulling the trigger of a gun "causes" a death. We need to know about the barrel of the gun, etc etc.

No... you're saying the equivalent of the bullet causes the death - without involving the gun, or the trigger, or the firer.

KBuck —Preceding unsigned comment added by KBuck (talk • contribs) 21:59, 6 January 2008 (UTC)

No, twice. Saying "the bullet causes the death" witout mentionning gun, etc, is perfectly viable. The text you added to the article is equivalent to stating that "the bullet doesn't cause the death" which as I say goes too far. Secondly, you are asserting in (1) above that The later state is the product of the WHOLE of the earler state - which is of course perfectly true - not some small PART of the earlier state you declare important - which would include pulling the trigger. To take this argument forward, you need to rephrase your (1) above William M. Connolley (talk) 22:14, 6 January 2008 (UTC)

We could say the bullet caused the death - if we're selecting bullet+death as the system (when of course we all know that is not the whole picture) - to say the butterfly causes the hurricane is like saying that the birth mother of the firer of the gun when giving birth caused the death.

The argument has plausibility because a butterfly has been chosen and we can imagine that that butterfly has 'free will'. Lets clear the waters by considering that cubic inch of atmosphere. (The butterfly is no less an arbitrary definition: it is continuous with the atmosphere and you cannot define for example which atoms of oxygen being breathed are part of the butterfly and which are not - except through your own arbitruary definition)

So - we have a given state of atmosphere involving the condition of that cubic inch - which gives rise to that later state. Now, it is true that had that particular cubic inch differed then the later state would not have arisen. But that is true of EVERY and ANY cubic inch you choose.

The hurricane is equally the product of EVERY butterfly-sized element from which it arose. Just because you arbitrarily abstract out one part of the atmospheric system and point to it as the cause doesn't make it so. The end result depends on all the other elements in the initial state also being in the particular states in which they are in - not just the butterfly.

KBuck —Preceding unsigned comment added by KBuck (talk • contribs) 22:34, 6 January 2008 (UTC)

It would be nice if your could sign your posts, and learn about indentation. Past that, The hurricane is equally the product of EVERY butterfly-sized element from which it arose - true; Just because you arbitrarily abstract out one part of the atmospheric system and point to it as the cause doesn't make it so - fairly true; but its false to try to turn it around and assert that the butterfly is *not* the, or a, cause. I don't know if you're still defending your para (1) above. Shall we just ignore it instead? William M. Connolley (talk) 22:46, 6 January 2008 (UTC)

"A cause", as in part of the prior history of the system is totally different from "the cause". "The" cause is what's being put forward. The notion captures the public imagination - but is nonsense.

I liked the para I wrote... but then I would... I'm not going to fight over it - the explanations here were more effort! Really I should be doing something else. This is just an avoidance strategy. :-) I'll look at signing and indents when I need another one.

KBuck —Preceding unsigned comment added by KBuck (talk • contribs) 22:56, 6 January 2008 (UTC)

I can think of other big problems with the illustration... the extra energy input from a butterfly's wing flag doesn't equal that of a hurricane - for the cause of the weather event you have to look to the energy source. Someone who knows about climate science/models could perhaps give the article the practicals on the real possible effects of tiny variations in air pressure upon the weather system over time.

KBuck —Preceding unsigned comment added by 62.252.240.53 (talk) 10:35, 7 January 2008 (UTC)

Really we've been having the old camel straw argument - you've been blaming the final straw (surmised variation) for breaking the camel's back - I've been saying the cause is the whole load. Its a way of looking at it to say the final straw is responsible - but equally we could pick any other added. It ignores the context, and involves dubious notions of separating out the butterfly from the rest of the system and it's history (our mental construct) and that of contingency, possibly free will, in the universe. —Preceding unsigned comment added by 62.252.240.9 (talk) 12:31, 7 January 2008 (UTC)

Good point. Not only is our tendency to say what role the butterfly's wings could or could not have played a reflection of human value judgments, the whole idea that we can easily determine where the butterfly ends and its "surroundings" start is not easy to do without a theory of causal judgments. That might not be so bad, except that what counts as a "cause" and "effect" depends a great deal on what type of claim is being made, and a tradition going back at least to Hume suggests that it would be more accurate to talk about correlation and conjunction, not "causes."

In that sense, it's interesting that there is a great deal of attention to hurricanes and tornadoes (which have significant human consequences) and less to things like whether the wings' motion disturbs leaves on a plant, and so on. It might be true that the average reader misunderstands the butterfly effect; but it seems equally likely that most any definition of it will be narrowed so that it is easier to grasp (e.g., the way that the causal transition from the insect kingdom is made through aerodynamics, to meterology, as though there were no "stops" in between). C d h (talk) 03:15, 30 April 2008 (UTC)

I don't believe that people can read all this and still be so wide of the mark with the very essence of what the butterfly effect is all about. It is not meant to be taken literally, it is not implied that a single butterfly in and of itself can cause or not cause a tornado. The notion being conveyed is that the end result of some systems depends on a virtually infinite number of infinitesimally small variations, of which the metaphorical butterfly is merely one. In other words, the causes and effects of some systems are so complex that they cannot be modelled and therefore we may as well state that for all practical purposes such events happen without any reason. 79.77.32.208 (talk) 20:11, 6 May 2008 (UTC)