[comp.graphics] Definition of a Fractal?

ron@oscvax.UUCP (08/27/87)

The Ontario Science Centre is having a temporary exhibition in the fall
and one part of this exhibition deals with fractals.  As the show grows
closer I am often asked for a definition of a fractal.  My attempts
usually result in a lot of hand-waving and showing of pretty pictures. 
I usually end up giving properties that are common among a lot of
fractals such as highly irregular, self-similar, non-differentiable,
etc.  The definition that Mandelbrot uses in The Fractal Geometry of
Nature is:

"A fractal is by definition a set for which the Hausdorff Besicovitch
dimension strictly exceeds the topological dimension".

Needless to say this is not something we can use on the unsuspecting
public.  Does anybody out there want to try and come up with a
definition for a fractal which is understandable to quasi-intelligent
people? 

Thanks for listening,
-- 
Ron Janzen
Ontario Science Centre, Toronto
...!{allegra,ihnp4,decvax,pyramid}!utzoo!oscvax!ron

lewisd@homxc.UUCP (D.LEWIS) (09/02/87)

In article <508@oscvax.UUCP>, ron@oscvax.UUCP (Ron Janzen) writes:
> I usually end up giving properties that are common among a lot of
> fractals such as highly irregular, self-similar, non-differentiable,
> etc.  The definition that Mandelbrot uses in The Fractal Geometry of
> Nature is:
> 
> "A fractal is by definition a set for which the Hausdorff Besicovitch
> dimension strictly exceeds the topological dimension".
> 
> Needless to say this is not something we can use on the unsuspecting
> public.  Does anybody out there want to try and come up with a
> definition for a fractal which is understandable to quasi-intelligent
> people? 

I think that you're on the right track with "self-similar."  Perhaps
also try "partial-dimensional."  Make the comparision between a straight
line and a space-filling curve, perhaps, and point out the similarity
of fractals to such curves.

-- 

David B. Lewis    {ihnp4!}homxc!lewisd
201-615-5306 Eastern Time, Days.