X-Message-Number: 27583
References: <>
From: Peter Merel <>
Subject: Mythdirection
Date: Mon, 6 Feb 2006 15:05:45 +1100

RBR writes,

>> Then the volume of matter enclosed in a region of space R1 at time
>> T1 has the same identity as the volume of matter enclosed in a
>> region of space R2 at time T2 if the following hold true: 1. P(R1,  
>> T1);
>> 2. P(R2, T2); and, 3. There exists an infinitely continuous  
>> function R(T) defining a
>> time-dependent region for which R(T1) = R1, R(T2) = R2. This IS  
>> the definition
>> of identity in a material world such as ours.
>
> At some point during the
> melting of the ice sculpture, and certainly when nothing remained
> but a pool of water, the predicate 'P(x) = x is an ice sculpture of
> a Nordic beauty' ceased being true, and the statue stopped existing.

The assertion that these particular water molecules form a sculpture  
of a nordic beauty relies on aesthetics. I continue to assert same  
because, as an artist, I know my work is imperfect if it does not  
reflect the evanescence of beauty. My guinean client, focused on mere  
pecuniary matters, disagrees. Your "concrete" definition admits this  
ambiguity.

Let's suppose you side with my client on contrarian grounds. Now let  
7 years pass. Every atom of my body has been replaced through the  
normal function of metabolism. By your definition and aesthetics, you  
and my guinean client must admit I have ceased to be the sculptor of  
the vanished nordic beauty. By your lights the sculptor's identity,  
along with that of the sculpture, has flushed into the ocean.

I'm disappointed you didn't jump at the live lady vs. live worms bit.

> The definition I gave used the criterion of infinite continuity
> because no one knows yet whether space is quantized (the only thing
> we know is you can't measure distances beyond a certain scale).
> It's not required but it's much cleaner than an alternative
> definition that assumes the discretness of spacetime.

"Time-dependent region" is likewise an anthropomorphism. Cf. Bell,  
Aspect, and so on.

> And if by simulating a brain, you mean predict short-term
> macroscopic observables (such as, how long it would take your brain
> to cool to room temperature if it were ripped from your skull by an
> axe-wielding Pomeranian), then yes, we can 'simulate' a brain, in
> the same way we can 'simulate' a gas.

We have no adequate definition of general intelligence, so we do not  
know that general intelligence requires all those quatillions of  
flibbertygibbets more than a jumbo requires bird-feathers.

> We are lumps of matter.

I am demonstrably the entire universe and everything in it. Any  
"lumpish" distinction you prefer ignores the fundamental  
interconnection of all life. And Bell, Aspect, and so on.

We are an entropic process evolving by the opportunistic adaptation  
of non-local constituents, lumpy or otherwise no one can say, as  
generated by local enthalpy.

Or some other sentence with the word "process" in it.

> Believing you're right is dangerous because it means you will use
> any means (fallacious reasoning, ad hominems, etc.) to justify the
> end (converting the masses).

Poor masses, my heart goes out to them. Oh, wait, are we still  
speaking in the "lumps of matter" sense? Poorly huddled masses?

> Like most people familiar with IT, you like to think of the neuron
> as having N electrical inputs, and 1 electrical output. Nothing
> could be further from the truth.

In a sentence you failed to quote I said the little buggers do other  
things. As for connectionism I refer you to http://www.cryonet.org/ 
cgi-bin/dsp.cgi?msg=25557. Until someone bests my butterfly, I'm a  
diehard skeptic.

The point of my Moravec is to counter your "

> A virtual brain being simulated on a computer is no more self-aware
> and capable of subjective feeling than the hunk of plastic I'm
> typing this message on now.

". Since you seem to be backsliding on that point, let's agree not to  
worry about it.

> This same problem [combinatorics induced by simultaneity]
> is everywhere in physics (pretty much anywhere
> you see a differential equation).

To solve it would require a new paradigm of computation. The universe  
seems quite obliging in this regard; quantum computers will crunch  
the snot out of the most explosive Turing combinatorics. Many things  
come to pass.

> until someone can demonstrate otherwise, there's no
> reason to believe that limit is anywhere near ONE TRILLION (which
> is the number of neurons you need to simulate if you want something
> like a human brain).

Ever hear Bucky Fuller's rant on Pi? Basically Bucky looking at the  
white wake of a battleship realized that it was caused by TRILLIONS  
of round bubbles. He felt the intuition that it would be so  
uneconomical for the universe to use an infinite precision Pi to  
calculate the correct surface for each bubble that the proposition  
that it does so is not supportable. "The universe," he said, "isn't  
using Pi!"

Likewise it seems intuitive to suggest we need not simulate one  
trillion neurons to make something with the capability of a human  
brain. Intuition for or against not being bankable it behooves us to  
wait and see.

> I could go on and on, describing all of the
> factors that influence the shape of the 'neural function'; pointing
> out that these are changing with time, and, as with all truly
> interesting problems, are self-dependent. But I won't.

I could go on counting the fractal branch points of a feather until I  
was blue in the gizzard. Or I could refer myself to Barnsley's work  
on Iterated Function Systems and observe that a feather, fernlike,  
can be generated by an IFS represented by less information than this  
sentence. Similarly L-systems, and so on and so forth. Apparent  
complexity of function does not entail necessary complexity of  
function. The fact is, despite the handwaving, neither of us can make  
a claim here without a proof. And for intelligence we not only lack  
proofs - we most likely lack an adequate frame within which to  
express 'em as well.

> You want to put your money where your mouth is?

That's the Limpinwood X-Prize.

> The burden of proof is on you. You want me to believe, in spite of
> the evidence to the contrary, that neurons can be modeled and
> simulated using neat little mathematical equations?

No. I don't know that. I've only suggested you don't know the  
contrary. That said, yes, one of my pet projects is a dram o' math  
that might enable AI. Or it might not. I don't claim a thing without  
some kind of proof. Do you?

Peter Merel.

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=27583