Since geeky science questions seem to be today's fashion...
-
Not so. You always have the outside edges of the starting triangle; those remain the same size. What you are adding are the edges created by cutting out the middle triangle, which are half the size of the triangle's outer edges.
yes the number of edges increases; none of the new edges is larger than the three original ones. Or are we discussing something other than the Sierpinski Triangle now? You said the shape remained finite as it was limiting itself to the bounds of the original triangle. :)
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
-
Hmm. When you start with a finite triangle, and all you do is cut some edges in half, none of the edges will ever grow, let alone grow to infinity; the perimeter yes, the edges no. :)
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
Hmmm, after drawing it out, I think you are correct. While the edges get smaller, all the angles stay the same. They never really blend into a single edge. On the other hand, I still haven't been able to poke any holes in my circle fractal proposal. Feel free to do that for me. ;P
-
Hmmm, after drawing it out, I think you are correct. While the edges get smaller, all the angles stay the same. They never really blend into a single edge. On the other hand, I still haven't been able to poke any holes in my circle fractal proposal. Feel free to do that for me. ;P
aspdotnetdev wrote:
I still haven't been able to poke any holes in my cirlce fractal proposal
The way I see it, it has: - no straight parts at all; - a single edge, infinitely long; - a real, non-zero, area (using the even/odd rule for inside/outside), intuitively I would say half that of the original circle. So I wouldn't call it a fractal, I wouldn't call it a solution to the problem in the OP, and frankly I wouldn't call it very pretty either. It is original though, I haven't seen it before. :)
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
-
I'm surprised that no one has suggested fractals. The one I had in mind is called the Sierpinski Triangle. Observe: 1. Draw an equilateral triangle. 2. For each triangle, find the mid-point of its sides and draw lines connecting them, creating four triangles. 3. Remove the triangle in the middle, leaving three equilateral triangles connected at their vertices, each having three edges that are half of the starting triangle. 4. Go to step 2. The triangle starts with three edges of length x, so its total perimeter is 3x. After the first iteration, the shape has nine edges -- three on each of the three triangles -- each of which have a length of x/2, meaning the shape's total perimeter is 9x/2, longer than what we started with. After the first iteration, the area is 3/4 what it was before. As you continue with more iterations, the number of edges increases without bounds, and so does the shape's perimeter. As the number of iterations approaches infinity, so do the number of edges and, consequently, the length of its perimeter. Also, each iteration decreases the area geometrically: as the number of iterations approaches infinity, the area bound by the perimeter approaches zero. The shape itself never exceeds the bounds set by the starting triangle, which makes it finite. With a variant called the Sierpinski Carpet, you start with a single square, divide it into nine squares, remove the center one and repeat. Again, the number of edges and the perimeter approach infinity while the area bound by the perimeter approaches zero. There are also 3-d versions of these shapes, called sponges, which start with a tetrahedron and a cube respectively. Added: Oops, a bit of a screw up. The number of edges is doubled, not tripled; I was counting the starting edges of the triangle twice. The increase in the length of the perimeter is still 9x/2, however, as you have the three starting edges (x + x + x = 3x or 6x/2) plus the three edges of the now empty center triangle (x/2 + x/2 + x/2 = 3x/2)
modified on Thursday, May 6, 2010 8:31 PM
Whoa, whoa wait a second. Peano curve[^] beats serpinski triangle for your criteria by a wide margin.
I can imagine the sinking feeling one would have after ordering my book, only to find a laughably ridiculous theory with demented logic once the book arrives - Mark McCutcheon
-
aspdotnetdev wrote:
I still haven't been able to poke any holes in my cirlce fractal proposal
The way I see it, it has: - no straight parts at all; - a single edge, infinitely long; - a real, non-zero, area (using the even/odd rule for inside/outside), intuitively I would say half that of the original circle. So I wouldn't call it a fractal, I wouldn't call it a solution to the problem in the OP, and frankly I wouldn't call it very pretty either. It is original though, I haven't seen it before. :)
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
Luc Pattyn wrote:
a real, non-zero, area (using the even/odd rule for inside/outside)
Are you saying that the outer crescent (for want of a better term) is solid, the one inside that is not, the one inside that is, and so on? Not what I was thinking... I was thinking more like a spool of wire... or a hose that you are wrapping up by looping it around your hand and the part of the hose you've already wrapped. Suppose the hose is infinitely thin and you are wrapping in reverse (starting with the largest circle).
Luc Pattyn wrote:
So I wouldn't call it a fractal
By what definition? Start by viewing the whole thing: you see a circle with smaller nested circles. Now, zoom in with top of the "camera" staying focused on the top of the circle. You now see a circle with smaller nested circles. Continue zooming in and it looks pretty much the same, no matter how far you zoom in.
Luc Pattyn wrote:
I wouldn't call it very pretty either
Pft, whatever... I'm going to draw it and make it my desktop background. ;P
Luc Pattyn wrote:
I wouldn't call it a solution to the problem in the OP
Seems to satisfy all the requirements to me. How about a spiral (like the kind you get hypnotized with)? Only with a different function to determine the rate of shrinkage. If you still aren't convinced, just look here and I think you'll come to your senses eventually. You will come to your senses... you will come to your senses... you will come to your senses... ;)
-
I don't think you need polish, a little push will suffice, hence: The integral of
1/xisln(x) + some constant, and your function is symmetrical around the first diagonal, so the integral from 1 to infinity would cover one quarter of the total area (ignoring signs), and that quarter is infinite as it equalsln(infinity)Sorry, I can't store it without damaging it. :)Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
We're gonna need a bigger box.
-
Luc Pattyn wrote:
a real, non-zero, area (using the even/odd rule for inside/outside)
Are you saying that the outer crescent (for want of a better term) is solid, the one inside that is not, the one inside that is, and so on? Not what I was thinking... I was thinking more like a spool of wire... or a hose that you are wrapping up by looping it around your hand and the part of the hose you've already wrapped. Suppose the hose is infinitely thin and you are wrapping in reverse (starting with the largest circle).
Luc Pattyn wrote:
So I wouldn't call it a fractal
By what definition? Start by viewing the whole thing: you see a circle with smaller nested circles. Now, zoom in with top of the "camera" staying focused on the top of the circle. You now see a circle with smaller nested circles. Continue zooming in and it looks pretty much the same, no matter how far you zoom in.
Luc Pattyn wrote:
I wouldn't call it very pretty either
Pft, whatever... I'm going to draw it and make it my desktop background. ;P
Luc Pattyn wrote:
I wouldn't call it a solution to the problem in the OP
Seems to satisfy all the requirements to me. How about a spiral (like the kind you get hypnotized with)? Only with a different function to determine the rate of shrinkage. If you still aren't convinced, just look here and I think you'll come to your senses eventually. You will come to your senses... you will come to your senses... you will come to your senses... ;)
OK, I understood your concoction as a two-dimensional drawing where you have N "top-connected" circles of decreasing diameter (with N increasing without bound), and you travel to the next circle every time you reach the top, until you reached the smallest one, then you step to the outer one again. So that is a closed line, there is nothing to zoom that would keep the overall impression, and the area is a half circle. Not a fractal. If you want to visualize it as a spiral, i.e. each next circle moves you a bit in the third dimension, then you have somewhat of a fractal effect as you can move forward over the pitch of the spiral, and zoom in a bit to compensate for the decreasing diameter. But now it is just a spiral, it spans an infinite z-axis. So it is not contained in a finite space. (In fact it resembles a worm hole in Stargate-1). All my senses and I agree yours is not a space-limited fractal, and not the right answer to the OP. But I agree you might still like it as a wall paper. You, not me. :laugh: This[^] might offer some consolation.
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
-
Whoa, whoa wait a second. Peano curve[^] beats serpinski triangle for your criteria by a wide margin.
I can imagine the sinking feeling one would have after ordering my book, only to find a laughably ridiculous theory with demented logic once the book arrives - Mark McCutcheon
In terms of area, remember that
null != 0;:laugh: -
OK, I understood your concoction as a two-dimensional drawing where you have N "top-connected" circles of decreasing diameter (with N increasing without bound), and you travel to the next circle every time you reach the top, until you reached the smallest one, then you step to the outer one again. So that is a closed line, there is nothing to zoom that would keep the overall impression, and the area is a half circle. Not a fractal. If you want to visualize it as a spiral, i.e. each next circle moves you a bit in the third dimension, then you have somewhat of a fractal effect as you can move forward over the pitch of the spiral, and zoom in a bit to compensate for the decreasing diameter. But now it is just a spiral, it spans an infinite z-axis. So it is not contained in a finite space. (In fact it resembles a worm hole in Stargate-1). All my senses and I agree yours is not a space-limited fractal, and not the right answer to the OP. But I agree you might still like it as a wall paper. You, not me. :laugh: This[^] might offer some consolation.
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
Luc Pattyn wrote:
OK, I understood your concoction as a two-dimensional drawing where you have N "top-connected" circles of decreasing diameter (with N increasing without bound), and you travel to the next circle every time you reach the top
Yep.
Luc Pattyn wrote:
until you reached the smallest one, then you step to the outer one again
There is no "smallest" one. It keeps going, forever.
Luc Pattyn wrote:
there is nothing to zoom that would keep the overall impression
I used my excellent skills as an artist to make this rendition of what I was thinking. Suppose you started out zoomed to view the full shape. Then, you zoom so that you can only view the part of the shape composed of light grey circles. Then you keep zooming in that fashion. You'll always see circles within circles, all intersecting at the top of the view. It is this self-similarity that I used to define this as a fractal.
Luc Pattyn wrote:
not the right answer to the OP
Nonsense!
Luc Pattyn wrote:
This[^] might offer some consolation.
Sorry it took so long to respond... I just woke up from a seizure induced by that crazy image. ;P
-
Luc Pattyn wrote:
OK, I understood your concoction as a two-dimensional drawing where you have N "top-connected" circles of decreasing diameter (with N increasing without bound), and you travel to the next circle every time you reach the top
Yep.
Luc Pattyn wrote:
until you reached the smallest one, then you step to the outer one again
There is no "smallest" one. It keeps going, forever.
Luc Pattyn wrote:
there is nothing to zoom that would keep the overall impression
I used my excellent skills as an artist to make this rendition of what I was thinking. Suppose you started out zoomed to view the full shape. Then, you zoom so that you can only view the part of the shape composed of light grey circles. Then you keep zooming in that fashion. You'll always see circles within circles, all intersecting at the top of the view. It is this self-similarity that I used to define this as a fractal.
Luc Pattyn wrote:
not the right answer to the OP
Nonsense!
Luc Pattyn wrote:
This[^] might offer some consolation.
Sorry it took so long to respond... I just woke up from a seizure induced by that crazy image. ;P
aspdotnetdev wrote:
I just woke up from a seizure induced by that crazy image
And you make even more sense than before. Yes, I stand corrected, your 2D image has some fractal behavior; it is a bit special as it seems to require one looks through a circular hole; that way you can hide the larger circles entirely while zooming in, something a square view fails to do for moderate zoom levels. The area problem remains; the odd/even rule still seems to lead to a half circle area. What does Gregory say on the subject? PS: excellent artistic skills indeed; why didn't you fill the interior to settle the area issue too? :)
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
-
aspdotnetdev wrote:
I just woke up from a seizure induced by that crazy image
And you make even more sense than before. Yes, I stand corrected, your 2D image has some fractal behavior; it is a bit special as it seems to require one looks through a circular hole; that way you can hide the larger circles entirely while zooming in, something a square view fails to do for moderate zoom levels. The area problem remains; the odd/even rule still seems to lead to a half circle area. What does Gregory say on the subject? PS: excellent artistic skills indeed; why didn't you fill the interior to settle the area issue too? :)
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
Luc Pattyn wrote:
the odd/even rule
Luc Pattyn wrote:
half circle area
Luc Pattyn wrote:
fill the interior to settle the area issue
Still no idea what you are talking about there. It is an infinitely thin/long string wrapped in a bunch of circles (perhaps "loops" would be a better term, as the "circles" aren't filled in). There's no surface, so there's no area! You'll send me to the looney bin, you will.
-
In terms of area, remember that
null != 0;:laugh:Gregory.Gadow wrote:
In terms of area, remember that null != 0;
which metric? ha! Infinite length of edges, zero area and covers an n-dimensional volume. For an added bonus it's used to show that n-intervals have the same cardinality as the 1-interval. What's not to love? Your fractal only has an area that tends to zero, mine has zero volume before the limit :)
I can imagine the sinking feeling one would have after ordering my book, only to find a laughably ridiculous theory with demented logic once the book arrives - Mark McCutcheon
-
That would be a fractal, such as this Sierpinski triangle[^]. ADDED Although not many would agree they have 2 or 3 (or any integer) number of dimensions... :)
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
modified on Thursday, May 6, 2010 6:06 PM
My fractal has less area than yours. neener neener neener.
I can imagine the sinking feeling one would have after ordering my book, only to find a laughably ridiculous theory with demented logic once the book arrives - Mark McCutcheon
-
Luc Pattyn wrote:
the odd/even rule
Luc Pattyn wrote:
half circle area
Luc Pattyn wrote:
fill the interior to settle the area issue
Still no idea what you are talking about there. It is an infinitely thin/long string wrapped in a bunch of circles (perhaps "loops" would be a better term, as the "circles" aren't filled in). There's no surface, so there's no area! You'll send me to the looney bin, you will.
So now it is a spiral in 2D, it no longer is a collection of circles. Just an infinite line, curled rather than straight? That's a bit disappointing...
aspdotnetdev wrote:
You'll send me to the looney bin, you will.
We could organize a geeky science home party then. :laugh:
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
-
In terms of area, remember that
null != 0;:laugh:Gregory.Gadow wrote:
null != 0;
in my world, that does not even compile. :)
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
-
My fractal has less area than yours. neener neener neener.
I can imagine the sinking feeling one would have after ordering my book, only to find a laughably ridiculous theory with demented logic once the book arrives - Mark McCutcheon
maybe your fractal is broken? :)
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
-
Luc Pattyn wrote:
the odd/even rule
Luc Pattyn wrote:
half circle area
Luc Pattyn wrote:
fill the interior to settle the area issue
Still no idea what you are talking about there. It is an infinitely thin/long string wrapped in a bunch of circles (perhaps "loops" would be a better term, as the "circles" aren't filled in). There's no surface, so there's no area! You'll send me to the looney bin, you will.
It doesn't actually fill a volume unless you choose your radii carefully. For example if you pick r(i) = (1/2)i than you have finite sized gaps between each of the circles. You'd have to pick something like 1, 1/2, 1/3, 2/3, 1/4, 3/4, 1/5, ... so that it ended up filling the whole circle. Also without specifying the way that the radii decrease, you could end up with a finite length as opposed to an infinite one.
I can imagine the sinking feeling one would have after ordering my book, only to find a laughably ridiculous theory with demented logic once the book arrives - Mark McCutcheon
-
maybe your fractal is broken? :)
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
:rolleyes:
I can imagine the sinking feeling one would have after ordering my book, only to find a laughably ridiculous theory with demented logic once the book arrives - Mark McCutcheon
-
So now it is a spiral in 2D, it no longer is a collection of circles. Just an infinite line, curled rather than straight? That's a bit disappointing...
aspdotnetdev wrote:
You'll send me to the looney bin, you will.
We could organize a geeky science home party then. :laugh:
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles]
Prolific encyclopedia fixture proof-reader browser patron addict?
We all depend on the beast below.
Luc Pattyn wrote:
So now it is a spiral in 2D, it no longer is a collection of circles. Just an infinite line, curled rather than straight? That's a bit disappointing...
I'm glad we finally understand eachother. :rolleyes:
Luc Pattyn wrote:
We could organize a geeky science home party then.
I hear Weven is hosting these promotion parties. Yay, sounds like fun! ;P
-
It doesn't actually fill a volume unless you choose your radii carefully. For example if you pick r(i) = (1/2)i than you have finite sized gaps between each of the circles. You'd have to pick something like 1, 1/2, 1/3, 2/3, 1/4, 3/4, 1/5, ... so that it ended up filling the whole circle. Also without specifying the way that the radii decrease, you could end up with a finite length as opposed to an infinite one.
I can imagine the sinking feeling one would have after ordering my book, only to find a laughably ridiculous theory with demented logic once the book arrives - Mark McCutcheon
How about we say the distance the from the outer circle is a function of the angle. How about:
radius = 100 - 1/(1 + 1/(angle! + googleplex * ackerman(angle, angle)))
That ought to work. :rolleyes:
