SV: Capacity of the brain - The 'Point'

Chris Simmons cps102 at york.ac.uk
Mon Sep 20 13:51:18 EST 1999


On Fri, 17 Sep 1999, Ken Collins wrote:

> further note:
> 
> this no-'numbers', relative-length computational methodology is, BTW, why i
> commented, earlier in this thread, that in one second, the nervous system
> does more computation than that done by all of the [conventional]
> computation, whether done by human or machine, that've ever existed, or will
> ever exist.
> 
> of course, the no-'numbers' methodology can be simulated using numbers, but
> for each 'number' incorporated within such simulations the nervous system
> does an infinitely-large computation.
> 
> =everything= that nervous systems do is in Loving embrace with infinity.

Well, yes, that's conceivably true, but whether nature actaully managed to
produce indiscrete computers is another matter.  This is basically what
you seem to be saying - that neurons, since operating 'continuously' in
the sense of continuous maps are by default more powerful that digital
machines.

However, it seems to me that all this potential extra power is basically
unused by nature, since it's too hard to master; and thus fully digital
approximations would not be fundamentally less powerful.  Of course, this
*might* not be the case, but I at least consider it vastly unlikely.
[snip!]
> > > the 'no-numbers' stuff is from long-ago, in another online place... how
> > does
> > > computation happen without 'numbers'? via relative lengths... which is
[good grief, talk about wasting bandwidth... snip again!]

> > >
> > > K. P. Collins
> > >
> > >
> >
> >
> 
> 
> 
> 
[phew]




More information about the Neur-sci mailing list