Public
This movie shows the sense in which Julia sets are self-similar. The Julia set for a number z is the set of complex numbers that you can hit over and over with the function
f(x) = x^2 + z
and get a sequence of numbers that remains bounded. By definition, the Julia set gets mapped to itself by this function f.
In this movie, when it's not wiggling around, the black stuff is the Julia set for a number z equal to roughly 0.8 + 0.2 i. But it's animated: at time t, we see what happens when you take the Julia set and apply the function
f(x,t) = x^{2^t} + tz.
When t = 0 this function does nothing. By the time t = 1, this function equals
f(x) = x^2 + z
so it maps the Julia set into itself. And then the animation loops around!
Anders Kaseorg put this animated gif on Quora:
http://www.quora.com/Fractals/Why-are-Julia-sets-fractals
but I don't know where he got it. And by the way, what I'm calling the Julia set for the number z is technically called the filled Julia set for the function f(x) = x^2 + z. For more definitions and pictures, see:
http://en.wikipedia.org/wiki/Julia_set
#fractals
f(x) = x^2 + z
and get a sequence of numbers that remains bounded. By definition, the Julia set gets mapped to itself by this function f.
In this movie, when it's not wiggling around, the black stuff is the Julia set for a number z equal to roughly 0.8 + 0.2 i. But it's animated: at time t, we see what happens when you take the Julia set and apply the function
f(x,t) = x^{2^t} + tz.
When t = 0 this function does nothing. By the time t = 1, this function equals
f(x) = x^2 + z
so it maps the Julia set into itself. And then the animation loops around!
Anders Kaseorg put this animated gif on Quora:
http://www.quora.com/Fractals/Why-are-Julia-sets-fractals
but I don't know where he got it. And by the way, what I'm calling the Julia set for the number z is technically called the filled Julia set for the function f(x) = x^2 + z. For more definitions and pictures, see:
http://en.wikipedia.org/wiki/Julia_set
#fractals

I wish I knew what any of that meant, because that is far out.Jun 13, 2012
Nice! I've never thought of connecting two iterates by a homotopy.Jun 13, 2012
I think a more useful definition is to define the Julia set as the complement of the set of regular points. Regular points are intuitively the points for which a neighborhood is mapped smoothly along the whole orbit. Non-regular points are the ones for which the orbit gets chaotic. With this definition, the Julia set is only the boundary of the region of bounded orbits. See these great lectures by Milnor:
http://arxiv.org/abs/math.DS/9201272
Something really cool I realized recently about Julia sets, or at least about the generalizations you get by considering an arbitrary rational function, is that they really live on the sphere. Even cooler, some of these Julia sets fill the whole sphere and yield dense fractal patterns. Here is an example:
http://www.algorithmic-worlds.net/sphere/sphere.php?id=20111029
You need Java for the applet to work.Jun 13, 2012
Oops missed the last part of your post... apologies...Jun 13, 2012
No problem! I'm not an expert on Julia sets, so your comment had some news in it for me, even though I was deliberately working with the filled Julia set since it's quick to define and it's the black stuff in this movie.Jun 14, 2012
Can we make a similar animation for the Mandelbrot set?Jun 28, 2012
The Mandelbrot set is far more complicated: in particular, the region near the boundary of the Mandelbrot looks like all possible Julia sets. So I'd have no idea how to do something like this for the Mandelbrot set. See this:
http://www.quora.com/Fractals/Why-is-the-Mandelbrot-set-a-fractal
The combination of my answer and especially Anders Kaseorg's answer may give you some clues if you want to try this.Jun 28, 2012
Got around to trying it. It looks nice!
http://plus.google.com/110214848059767137292/posts/4UQ4WxJC8H9Sep 5, 2012
Add a comment...