On Wednesday, October 9, 2013 8:30:16 PM UTC-4, Liz R wrote:
> On 10 October 2013 13:03, Craig Weinberg <whats...@gmail.com <javascript:>
> > wrote:
>> On Wednesday, October 9, 2013 5:52:46 PM UTC-4, Liz R wrote:
>>> On 10 October 2013 09:47, Craig Weinberg <whats...@gmail.com> wrote:
>>>> It's not that computers can't do what humans do,* it's that they can't 
>>>> experience anything.* Mozart could dig a hole as well as compose 
>>>> music, but that doesn't mean that a backhoe with a player piano on it is 
>>>> Mozart. It's a much deeper problem with how machines are conceptualized 
>>>> that has nothing at all to do with humans.
>>> So you think "strong AI" is wrong. OK. But why can't computers 
>>> experience anything, in principle, given that people can, and assuming 
>>> people are complicated machines?
>> I don't think that people are machines. A machine is assembled 
>> intentionally from unrelated substances to perform a function which is 
>> alien to any of the substances. Living organisms are not assembled, they 
>> grow from a single cell. They have no unrelated substances and all 
>> functions they perform are local to the motives of the organism as a whole. 
> I believe that, at least in discussions such as this one, defining people 
> as machines has nothing to do with how or why they are constructed, and 
> eveything to do with ruling out any supernatural components. 

Right, but that's what I am saying is the problem. It would be like making 
generalizations about liquids based on water and saying that alcohol can't 
burn because it's a liquid. A machine and a person might both be able to 
say 'hello', but the machine was constructed by people who know what hello 
means, and the person knows what hello means because they were the ones who 
constructed the word. The word exists to serve their own agenda, not that 
of an alien programmer.


> Anyway, allow me to rephrase the question.
> I assume from the underlined comment that you think that strong AI is 
> wrong, and that we will never be able to build a conscious computer. How do 
> you come to that conclusion?

I guess that I came to that conclusion by first trying to exhaust the other 
alternatives and then by coming up with a way to make sense of awareness as 
what I call Primordial Identity Pansensitivity. This means that physics and 
information are incomplete reflections within sense rather than producers 
of consciousness. Physics is sense experience that is alienated by entropy 
(spacetime) and information is sense experience which has been alienated by 
generalization (abstraction). Information cannot be pieced together to make 
an experience. No copy can be made into an original. This is not because of 
some special sentimental feeling about consciousness, it's rooted in an a 
careful consideration of the number of clues that we have about perceptual 
relativity, authenticity, uniqueness, polarity, multiplicity, automaticity, 
representation, impersonality, and significance.

>> This is an even bigger deal if I am right about the universe being 
>> fundamentally a subdividing capacity for experience rather than a place or 
>> theater of interacting objects or forces. It means that we are not our 
>> body, rather a body is what someone else's lifetime looks like from inside 
>> of your lifetime. It's a token. The mechanisms of the brain do not produce 
>> awareness as a product, any more than these combinations of letter produce 
>> the thoughts I am communicating. What we see neurons doing is comparable to 
>> looking at a satellite picture of a city at night. We can learn a lot about 
>> what a city does, but nothing about who lives in the city. A city, like a 
>> human body, is a machine when you look at it from a distance, but what we 
>> see of a body or a city would be perfectly fine with no awareness happening 
>> at all. 
> Insofar as I understand it, I agree with this. I often wonder "how a load 
> of atoms can have experiences" so to speak. This is the so-called hard 
> problem of AI. It is (I think) addressed by comp.

If I'm right, then comp cannot address the hard problem. If we try to make 
it seem to address it, I think that it would have no choice but to get it 
exactly wrong. Comp fails because of the symbol grounding problem and the 
pathetic fallacy. It should be evident from Incompleteness, that no symbol 
can literally symbolize anything, and that all mathematical systems can 
only relate to isolated specifics or universal tautologies. Math cannot 
live because it can't change. It doesn't care. It doesn't know where it's 
been or where it's going. Comp is only one footprint of the absolute - the 
generic vacuum which divides experiences from each other. It misses 
presentation entirely, and so can only be a representation of 
representation...as Baudrillard would say, a Stage Four Simulacra:

"The fourth stage is pure simulation, in which the simulacrum has no 
relationship to any reality whatsoever. Here, signs merely reflect other 
signs and any claim to reality on the part of images or signs is only of 
the order of other such claims..."

This is why the Liars Paradox is not a paradox. These words are not 
literally meaningful. They require an interpreter who intends them to mean 
something that relates to their personal experience. To say "I am always 
lying." is not a paradox because all that can really be communicated is 
signs, not realities. "I am always lying" really means "Let us suppose that 
it is possible for someone to say that they are always lying." Every 
proposition is, from a realistic perspective, merely the proposition of a 
proposition. Truth cannot be communicated, it can only be understood. 
Communication can be used to help us understand truths, but it has no 
literal truth in it. We are all liars, since words are only words and not 
what we intend them to refer to.


You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to