[time 221] Re: [time 219] Re: [time 212] Re: [time 207] Re: [time 206] Observation and infinity

Ben Goertzel (ben@goertzel.org)
Sun, 11 Apr 1999 12:23:32 -0400

WE can never know whether the universe is finite or infinite, because we
are finite and
our knowledge is therefore finite

Hence, we are free to make whatever assumption makes our theories about our
domain of observation simpler!

To my way of thinking, assuming a finite universe makes my theorizing about
my domain
of observation simpler. But if an infinite universe winds up making our
theory come out simpler,
I won't cry too hard.

The reason to assume infinity in math is that it makes the formulas
simpler. Integration by parts
isn't as messy as summation by parts. Etc.


At 11:36 PM 4/10/99 -0400, Peter Wegner wrote:
>Stepehen and Ben
> I hesitate to come in in the middle of the discussion about finiteness
>versus infiniteness of the universe.
>But I have a preference for an infinite model, at least from the subjective
>viewpoint of observers, for the following kinds of reasons.
>1. From the viewpoint of general principles of relativity, observable
>properties should not depend on whether you are near the edge of the
>universe or near the center.
>If the observer cannot detect the edge of the universe this suggests an
>infinite model, at least from the subjective viewpoint of any observer.
>2. I prefer to think of the universe as an open rather than a closed system
>in that any part can be subject ot forces from unknown parts.
>An open universe is subject to nondeterministic external forces, while a
>closed universe can be modeled at some level in a deterministic
>(algorithmic) way.
>The existence of a closed deterministic universe seems to violate a general
>relativity principle.
>Openness is better matched by an infinite than a finite universe.
>This notion of openness is like that in topology.
>Topology allows us to clearly see that open sets can be finite, for example
>the open unit sphere, while still having the property that the complete set
>cannot be effectively defined.
>3. Mandelbrot sets are an attractive model for an infinite universe.
>This would allow a principle of relativity for scale, in that structures
at any
>particular scale would replicate themselves at both lower and higher scales.
>Some of Eddington's work suggests that the scale factor for replication
>might be about 10**80.
>If the universe has relativity of scale like a Mandelbrot set it is an
>infinite universe.
>Ben, has there been work on modeling the universe in terms of Mandelbrot sets?
>Are there models that focus especially on open versus closed systems?
> peter

This archive was generated by hypermail 2.0b3 on Sun Oct 17 1999 - 22:31:52 JST