In a message dated 05/18/2000 1:41:52 AM Pacific Daylight Time, 

> Actually, James, I read something wild into your sentence, interpreting "
> competition" as the
>  selection mechanism. That does strike me as more promising than Occam's 
> razor. I am interested
>  in pursuing this line of thought as a means of selecting not only ideas 
> perceived to be correct,
>  but also universes.
>  Anthropic compatibility has traditionally been the criterion to select 
> observed universes. It
>  does not appear feasible to apply this analogously to ideas, though. But 
> perhaps white rabbits
>  are naturally disfavored in this scheme?
>  Higgo James wrote:
>  > It seems to me that a good way of selecting one idea over is [sic] 
> competition is
>  > Occam's razor
>  >
Great Fred, I had exactly the same thought. The selection of ideas, just like 
the selection of life forms, does not have to follow Occam. Otherwise we 
would still be slime in the mud.

Anthropic selection is yet another kind of selection. 

While natural selection (selection of the individual by the world) is 3rd 
person effect, anthropy (selection of the world by the individual) is a 1st 
person effect. They are identical except that the frame of reference is 

Interestingly, these processes seem to be acausal: the fittest survives 
because it does. We see the world we see because we are alive to see it. 

What I mean is that none of these processes involve a "force" of nature such 
as gravitation or electromagnetism. They operate more at the logical level.  
Can the conventional physical forces be explained by means of logic?  Is the 
foundation of the world actually logical? This, I think,  is the road that 
the computationalists are attempting to follow.

This whole issue involves measure, probabilities and wabbits, so it is very 
controversial in this group.

I stongly believe that Occam does not operate at the anthropic level, just as 
it does not operate at the Darwinian level, and that computation of measures 
using Occam approach is highly suspect.


Reply via email to