Talk:Borromean rings/Archive 1

Mobius Strip
Can this arrangement be twisted about into a Mobius Strip? THAT would be weird. Same surface continuity, no two rings connected, yet "one" complex. I can't get my mind around that! bt -- unsigned comment by anonymous IP 68.102.13.50, 06:48, 26 August 2006

clarification
Why wouldn't you be able to form this figure from regular circles? Or should it say from two-dimensional figures, as the linkages require overlap? -- nae'blis (talk) 20:27, 1 February 2006 (UTC)


 * No -- you can't do it with exact geometric circles. Take a close look at the picture. Michael Hardy 21:19, 1 February 2006 (UTC)
 * I can assure you that I designed my tattoo using geometric circles as the basis - can you explain that to me again? Is the 2.ε-dimensional overlap the 'impossibility', or the shape of the ellipse/circle? -- nae'blis (talk) 23:28, 1 February 2006 (UTC)


 * I didn't mean you can't make 2-dimensional pictures of it with exact circles. In the first place, any two of the circles would have to be in two different planes; otherwise they would have common points rather than being linked.  That means we need to embed them in a three-dimensional space.  I was speaking of the actual circles, not of pictures of them.  So is this article. Michael Hardy 23:46, 1 February 2006 (UTC)
 * Thanks, I'll make a small clarification to the opening paragraph for those who aren't as adept at topology. -- nae'blis (talk) 15:12, 2 February 2006 (UTC)
 * OK, I've found a reference: B. Lindstr&ouml;m, "Borromean Circles are Impossible", American Mathematical Monthly, volume 98 (1991), pages 340&mdash;341. I'm going to add that to the article.


 * But anyway, it seems you had in mind 2-dimensional pictures in which the circles are perfect circles. No one has said that those are impossible, and you already see those in the article. Michael Hardy 23:59, 1 February 2006 (UTC)

I've not seen the Math Monthly article, but I first learned about it from a short note by Ian Agol. The proof is fairly simple but ingenious. I don't know if we need two references, but this has the advantage of being freely available through the Internet (instead of say, through JSTOR). --C S (Talk) 01:34, 2 February 2006 (UTC)

I made a variation of Borromean Rings with a tangle toy. Each ring is pringle shaped, though.

Recent edits
The recent edits have, in my opinion, been rather lacking. Poor wording and even misleading statements have been inserted. --C S (Talk) 20:33, 12 March 2007 (UTC)

I do like the introduction of gallery tags. Anyway, I changed some things. --C S (Talk) 00:04, 14 March 2007 (UTC)

Diagrams in "Mathematical properties"
Surely the "ellipses of arbitrarily small eccentricity" are actually in the disgram to the left rather than the right as mentioned in text. I'll edit it but feel free to correct me.--Andyk 94 10:07, 3 August 2007 (UTC)

Anyone want to explain why it keeps being changed back?--Andyk 94 02:46, 10 August 2007 (UTC)


 * The 2D picture (with the yellow circle) doesn't represent a geometric situation with either circles or ellipses. It shows the basic configuration but the circles cannot actually be made that way without bending them.  The 3D picture (with the green) shows realizable ellipses.  I hope that clears up your confusion.  --C S (talk) 13:02, 9 March 2008 (UTC)


 * I agree. Andyk is wrong.  The picture in the center is the one that shows how this link can be realized with ellipses.  The one on the left shows a link with circles that is not realizable in Euclidean space. Michael Hardy (talk) 13:45, 10 March 2008 (UTC)

Eccentricity is a measure of how much a ellipse deviates from being circular. An ellipse with arbitarily small eccentricity would have it's foci aribitrarily close to the center, so it would look circular--that's the point, right? This should be mentioned, and the picture reference removed.


 * I'm not sure what you are asking to have mentioned. The middle picture of ellipses demonstrates a feasible configuration.  The one on the left with circles is not feasible, even were you to replace them with ellipses, since the rings would have to bend out of the plane.  That's the point of the picture reference which you want removed.  Without reference to the middle picture, it would be very difficult to see how to realize the Borromean rings with ellipses of small eccentricty.    --C S (talk) 20:52, 26 June 2008 (UTC)

The phrase "arbitrarily small eccentricity" indicates an ellipse that is almost circular, IE the picture on the far left. Or, to put it another way (and I hope this clarifies more than confuses), it displays the attempt at using circles but, as pointed out in te text and above, these circles would have to be bent, or, to keep them as 2d shapes, stretched (and, I believe, rotated slightly) IE turned from circles into ellipses. I shan't re-edit it because we're just going back and forth here. But if I've made it clear to those who were removing my edits, it should be corrected imediately.--Andyk 94 (talk) 16:06, 27 June 2008 (UTC)


 * No, again, let me repeat. The ellipse picture is NOT the one on the far left.  I don't know what you mean by rotated slightly.  The obvious interpretation doesn't work.  I have no idea why you are latching onto the left picture like you are.  The way to have a configuration with ellipses is to use the middle picture, which I think you don't understand.  The eccentricities of those ellipses can in fact be made arbitrarily small.  There is no correction needed, although I'm starting to wonder from your confusion whether there should be some rewriting to make the points clearer.     --C S (talk) 19:04, 27 June 2008 (UTC)


 * Your misconception stems from being fixated on the leftmost picture with the circles, apparently because the middle picture doesn't look circular enough to you. The middle picture can in fact be made as nearly circular as you wish.  Do you see how it uses ellipses to realize the Borromean Rings (topologically the same, but looks arranged differently)?  Next, do you see how you can make the eccentricities smaller (but not all the same) realizing that configuration? (Note the picture is drawn in 3D perspective)  From that, you can easily tell why ellipses make the configuration possible.  Now, you insist that the leftmost picture is realizable by ellipses.  Why?  How does making the rings ellipses help avoid the "bending" problem?  Even if you try rotating each one this way or that, you can't avoid the bending.  And it's certainly not clear why using ellipses would help avoid it.  According to you, somehow this "rotating" doesn't work when they are circles but works when they are nearly circles.  BTW, I wrote the material, and what you are thinking is certainly not what I had in mind, and it's not correct either.    --C S (talk) 19:28, 27 June 2008 (UTC)


 * I'm claiming that the central picture shouldn't be described as containing "ellipses of arbitrarily small eccentricity". By "rotated slightly" I was refering to rotation out of the plane in which we are working, as suggested above by another user. I have latched onto the leftmost picture because that is the one that seems to be best described by the text that currently references the central picture. I'm not certain how the central picture differs from the leftmot picture, other than the spacial arrangement and ecentricity of the rings, the relationship of the rings with each other seems to be identical. In which case, the alteration of eccentricity along with this rotation out of the single plane would surely make the 2 diagrams equivalent?--Andyk 94 (talk) 00:34, 19 July 2008 (UTC)

Andyk, you know how to miss a point. The point is: (1) Borromean links cannot be realized by circles, whether coplanar or not, and (2) They can be realized by ellipses that are not exactly circular, and (3) No matter how close to being circular they are, as long as they're not exact circles, the ellipses, if suitably positioned, can still realize the Borromean links. You can't have a picture of particular "ellipses of arbitrarily small eccentricity". Any particular ellipses that are not perfect circles are NOT of "arbitrarily small eccentricity"; they have whatever particular eccentricity they have and it's not zero. The words "arbitrarily small" merely express the point identified as (3) above, somewhat less long-windedly, but in language so completely standard that everyone (except non-mathematicians) will know that's what it means. Michael Hardy (talk) 11:57, 19 July 2008 (UTC)
 * OK, just to clarify further: it's not just that you can't have a picture of particular ellipses of "arbitrarily small" eccentricity. It's that no particular ellipse can have "arbitrarily small" eccentricity.  Any particular ellipse has whatever particular eccentricity it has.  The point of the picture is that ellipses that are not circles can serve in that role.  The "arbitrarily small eccentricity" part means they can still serve no matter how close they are to perfect circles as long as they're still not perfect circles.  But since the picture's purpose is to illustrate that ellipses that are not exactly circular can serve, it has to make clear from their appearance that they are ellipses that are not perfectly circular. Michael Hardy (talk) 17:30, 19 July 2008 (UTC)

OK, now I've rephrased it for the benefit of those not fluent in mathematicians' linguistic conventions. Michael Hardy (talk) 17:35, 19 July 2008 (UTC)

Sorry Mike, I'm not sure if you realise your attempts to partonise me are actually re-afirming my point or not but I'd love to go over it again. 1) Borromean links cannot be realized by circles, whether coplanar or not. I know, however, if stretched infinitesimally and rotated infinitesimally (again out of the initial plane) then a set of circles become a set of ellipses with "arbitrarily small" eccentricity. Now, assuming that these defections (eccentricity and planar rotation) from the impossible case (ideal circles in a single plane) the diagram of apparently circular rings fits the description perfectly, so long as the defections are too small to be recognised by the naked eye (and indeed, possibly too small to be displayed in pixels). Assuming the viewer doesn't assume a physically impossible configuration is being displayed, (s)he would have to assume that this is the case.


 * I'm not attempting to patronize you; I'm attempting to get correct information here and in the article. The point of the second picture is that Borromean links can be realized by ellipses that are NOT circles.  The picture should make that clear by making it clear that the curves are ellipses that are NOT circles.  The second picture does that. The first picture fails to do that.  The words "arbitrarily small eccentricity" mean only that ellipses can still serve no matter how small their eccentricity, as long as it is not exactly zero.  It is unreasonable to ask for a picture of ellipses of arbitrarily small eccentricity, since there is no such thing as an ellipse of arbitrarily small eccentricity. Michael Hardy (talk) 20:55, 19 July 2008 (UTC)

I've found some Wikipedia pages that can help
I've found these two pages:


 * Arbitrarily large
 * Primes in arithmetic progression

The first addresses the issue directly, explaining how mathematicians use this terminology. The second explicitly says there are "arbitrarily long, but not infinitely long" arithmetic progressions of prime numbers.

To say that there are arbitrarily long arithmetic progressions of prime numbers DOES NOT mean that there is any particular arithmetic progression of prime numbers that is "arbitrarily long". There is no such thing. It means that no matter what length you pick, no matter how big, there are arithmetic progression that are at least that long (but still may be finite, and in this case are always finite).

That's just part of the standard jargon of mathematicians. Michael Hardy

I understand that, and I'm not claiming that the picture does fit the description of "ellipses with arbitrarily small eccentriity". I'm claiming the left most picture comes closest to displaying the notion that any finite eccentricity will suffice ie by using ellipses with incredibly small though not "arbitrarily small" itself(since, as you point out, to claim any particular quantity is arbitrary makes no sense).--Andyk 94 (talk) 02:04, 21 July 2008 (UTC)

The second picture illustrates the idea that it can be done with ellipses. The fact that they can have eccentricity as small as desired is probably not something that could be illustrated by a simple picture. The fact that ellipses differing only infinitesimally from circles cannot be distinguished from circles in an accurate picture presents a problem. Michael Hardy (talk) 03:43, 21 July 2008 (UTC)

Images added to French version of article
-- AnonMoos (talk) 08:44, 19 February 2008 (UTC)

Pargraph needs editing
"The Borromean rings give examples of several interesting phenomena in mathematics. One is that the cohomology of the complement supports a non-trivial Massey product. Another is that it is a hyperbolic link: the complement of the Borromean rings in the 3-sphere admits a complete hyperbolic metric of finite volume. The canonical (Epstein-Penner) polyhedral decomposition of the complement consists of two ideal octahedra." - what the hell does any of this mean? -- anon —Preceding unsigned comment added by 81.99.106.40 (talk • contribs)


 * I means something that can be understood by those who know algebraic topology and some advanced geometry. Maybe that should be stated at the beginning of the paragraph. Michael Hardy (talk) 17:04, 22 March 2008 (UTC)


 * The Massey product thing sounds like gibberish, I agree. But the other things are actually very down-to-earth statements.  --C S (talk) 21:01, 26 June 2008 (UTC)

It doesn't sound like gibberish; it sounds like something that hasn't been defined here in this article. It has a link. The article it links to is unfortunately pretty terse, though. Michael Hardy (talk) 12:02, 19 July 2008 (UTC)

Mathematics section error
When you look closely at the first picture it is not a Borromean ring at all, the blue ring is not connected to the other two as it is on top of the yellow and under the red, but not interlocked. If rings were placed in such a fashion as that one, the blue could easily be taken apart without breaking it, and the other two would remain locked, clearly not borromean fashion. —Preceding unsigned comment added by Pyrofyr (talk • contribs) 04:25, 7 February 2009 (UTC)


 * If you mean File:BorromeanRings.svg, no two rings are directly interlinked, and that's actually the point. Blue is "under" yellow, red is "under" blue, and yellow is "under" red.  The whole thing has perfect threefold rotational symmetry, so I'm not sure why you're singling out blue... AnonMoos (talk) 08:29, 7 February 2009 (UTC)

Photo of older depiction
Classic medieval depiction of the rings as literal rings... AnonMoos (talk) 18:27, 6 May 2009 (UTC)

"Borromean nucleus"
Apparently a term sometimes used by physicists: http://www.anl.gov/Media_Center/News/2004/PHY041029.html -- AnonMoos (talk) 18:52, 9 May 2009 (UTC)

Borromean castle
http://rfcwalters.blogspot.com/2009/06/borromean-rings.html claims the admission tickets to the Borromeo family castle show an incorrect picture of the rings. I haven't traced it carefully enough to spot the error, but if there really is one, it may be worth mentioning in the article. 67.122.209.126 (talk) 08:20, 28 June 2009 (UTC)


 * That a version of the crest containing another 3 ring configuration is on a ticket is not so interesting, particularly when the crest itself was depicted that way by the Borromeos themselves, including on many carved doors and such. the Borromean family used several versions of their crest, some of which were not Borromean (hm, that sounded a bit funny now that I read it...).  This was discovered by some topologist who examined all the carvings, depictions, etc. of the crest in the castle.  The actual Borromean ring configuration wasn't used in a substantial portion of all of them.  So the ticket can be considered to depict a "legit" crest, unless you want to take the position that many depictions used by the Borromeos is wrong.  --C S (talk) 04:27, 29 June 2009 (UTC)

Literary reference
In Heinlein's Space Cadet, the Borromean rings were used as an emblem. AnonMoos (talk) 21:02, 8 February 2010 (UTC) --

Above the stage at the far end were the three closed circles of the Federation - Freedom, Peace, and Law, so intertwined that, if any one were removed, the other two would fall apart.

Multiple Borromean rings - Discordian mandala
How many Borromean rings configurations does the Discordian mandala contain? In my understanding, each configuration consists of three of the five 9-sided 'rings'. There are ten ways of choosing three from a group of five so I suggest the mandala contains not five but ten configurations. Less theoretically: Red-Yellow-Green; Red-Yellow-Blue; Red-Yellow-Magenta; RGB; RGM; RBM; YGB; YGM; YBM; GBM. 221.243.91.229 (talk) 06:02, 6 May 2011 (UTC)


 * Sorry I didn't see your remark before, but I really don't think so; if you choose a non-adjacent configuration, then they are not Borromeanly-linked... AnonMoos (talk) 01:07, 7 August 2011 (UTC)

Proof?
Is it just me, or is the synopsis of the Lindström & Zetterström proof incomprehensible? It goes like, "if one assumes that circles 1 and 2 touch at their two crossing points" – OK, but what if one does not assume that? Is the touching here assumed to take place in the "link diagram"? Isn't that supposed to be a plane figure? Then how do spheres enter the picture? (I have to confess I don't find the full proof as given in the AMM particularly convincing either: the "lifting" operation is not well defined; for $C_{2}'$ to meet $C_{1}$ in two points, just a translation will not do in general; additionally a rotation is needed, which makes it unclear what the statement means that $C_{2}$ is lifted "away" from $C_{3}$.) --Lambiam 11:27, 31 January 2012 (UTC)

Clover loop
This loop/link was drawn from some rearrangment of the Zeppelin knot,, but also could seen as a figure of eight with a particular extra loop. Would this match any of the knot theory work?SignedJohnsonL623 (talk) 11:58, 27 February 2013 (UTC)


 * It's not Borromean/Brunnian because it only contains one loop (i.e. is not a "link" in knot theory terms). It appears to be a non-alternating 9-crossing knot.  You could run it through the "Knotscape" program to see what it translates to in terms of standard knot inventories (see http://katlas.math.toronto.edu/ etc.). AnonMoos (talk) 14:58, 28 February 2013 (UTC)

Two cirlces, one ellipse of arbitrarily small eccentricity
While it has been mentioned that the Borromean rings can be realized as a link using three ellipses with arbitrarily small eccentricity, it is actually the case Borromean rings can be realized with two circles and one ellipse of arbitrarily small prescribed eccentricity. Here is a rather canonical example. Consider the circles (with 0 < b < a):

A: y2/a2 + z2/a2 = 1;  x = 0 in the yz-plane

B: x2/b2 + z2/b2 = 1;  y = 0 in the xz-plane

and the ellipse

C: x2/a2 + y2/b2 = 1;  z = 0 in the xy-plane

Obviously, A & B do not intersect, for they meet the z-axis at points a units and b units from the origin respectively. A & C meet the y-axis at points a units and b units from the origin, respectively. And, finally, B & C meet the x-axis at points b units and a units from the origin, respectively. Since 0 < b < a, these points are all distinct, and since a & b can be chosen as close to each other as desired, the eccentricity (1 - b2/a2)1/2 of ellipse C can be as small as desired. Chuck (talk) 05:22, 28 July 2013 (UTC)

Lacan
Some reference, however fleeting, should be here.--Dnavarro (talk) 00:46, 9 November 2015 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 1 one external link on Borromean rings. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Added archive http://web.archive.org/web/20110604122749/http://video.google.com/videoplay?docid=-5851348533617330039 to http://video.google.com/videoplay?docid=-5851348533617330039

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

Cheers.—cyberbot II  Talk to my owner :Online 22:26, 26 February 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 4 one external links on Borromean rings. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20071019011652/http://www.popmath.org.uk/sculpture/pages/2creatio.html to http://www.popmath.org.uk/sculpture/pages/2creatio.html
 * Added archive https://web.archive.org/web/20040805210510/http://www.liv.ac.uk:80/%7Espmr02/rings/index.html to http://www.liv.ac.uk/~spmr02/rings/index.html
 * Added archive https://web.archive.org/web/20071025213451/http://www.popmath.org.uk/sculpmath/pagesm/borings.html to http://www.popmath.org.uk/sculpmath/pagesm/borings.html
 * Added archive https://web.archive.org/web/20071027132444/http://www.popmath.org.uk/sculpmath/pagesm/africa.html to http://www.popmath.org.uk/sculpmath/pagesm/africa.html

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.— InternetArchiveBot  (Report bug) 07:54, 6 November 2016 (UTC)

Editing needed
The first paragraph is mathematically incorrect at several points. In the first sentence, the word unconnected should be replaced by unlinked. The phrase "In other words" in the second sentence is not correct, since the second sentence is not equivalent to the first sentence. The fact that:  no two of the three rings are linked with each other as a Hopf link, is not particularly relevant and is awkwardly phrased. A suggestion would be to replace unconnected with unlinked in the first sentence, to delete the second sentence, and to add a final sentence to the paragraph: The Borromean rings are characterized by being the nontrivial 3-component Brunnian link with minimal crossing number.

I tried to edit the second sentence and it was immediately undone. The ensuing discussion convinced me that I had neither the time nor energy nor knowledge of wikipedia for carrying out the edit. — Preceding unsigned comment added by Jfdavis (talk • contribs) 21:47, 21 August 2019 (UTC)


 * I'm sorry, but most of the cultural prominence of the Borromean rings among the general public comes from the initially-surprising fact that no two rings are directly linked (Hopf-linked) but all three are collectively interlinked. Without this Brunnian property, the Borromean rings would be of little interest except to mathematicians -- like the Whitehead link or the 6-crossing links other than the Borromean rings (NONE of which have separate Wikipedia articles devoted to them).  Therefore the Brunnian nature of the Borromean rings MUST be mentioned in the lead section -- and this needs to be explained (we can't automatically assume that the general reader will know what "Brunnian" means)... AnonMoos (talk) 16:42, 23 August 2019 (UTC)


 * P.S. Here's how they're described in two math/geometry books for non-specialists which I happen to have conveniently at hand: Cundy & Rollett Mathematical Models (2nd edition) -- "The three rings shown in Fig. 52(d) are such that no two are linked together, but all three are; cutting any one frees the other two" (page 59). Alan Holden, Shapes, Space, and Symmetry -- "The rings are inseparable, but no two are linked together; hence cutting any one frees the other two" (page 182). AnonMoos (talk) 14:38, 29 August 2019 (UTC)

Borromean rings "are" the Ballantine logo? Yes and no.
Using Google images to search for photos of the Ballantine logo, it is clear that some of the images — like this one: https://i.etsystatic.com/11180086/r/il/e72dcd/1829839974/il_1588xN.1829839974_jg7g.jpg — are indeed the Borromean rings.

But some either show an over/under arrangement that is definitely not the Borromean rings, or else contain no over/under information.

So it is not accurate to just state that the Ballantine logo "is" the Borromean link.2601:200:C080:630:49F3:543E:6F88:5D6A (talk) 22:02, 10 January 2021 (UTC)
 * We need to base our content on published reliable sources, not on our own surveys of the results of Google image searches. Do you have a reliable source that notes the variation in logo topology? —David Eppstein (talk) 23:22, 10 January 2021 (UTC)

Borromean Moebius strips
I have made something like this:

Allegra Pstrocski (talk) 22:59, 9 February 2021 (UTC)
 * Unless you have published sources for the significance of this construction, we can't use it in the article; see WP:NOR. --David Eppstein (talk) 23:58, 9 February 2021 (UTC)


 * Anyway, they're not actually basic (half-twist) Moebius strips... AnonMoos (talk) 20:18, 13 February 2021 (UTC)