My take on why indeed the study of consciousness may not be as simple

[rocketdodger]

I suggest you tell Rocketdodger and Pixymisa that. They seem pretty confident.

I find it indicative of honesty you bring to these discussions that you would repeat the statement

As yet unanswered.

This is why it's called the Strong AI hypothesis (or the Church-Turing hypothesis, if you prefer).

while purposefully excluding the rest of the reply made by drkitten:

However, there's a lot of support for the idea that the human brain is Turing equivalent, both physical and mathematical. The physical support comes from the observation that everything that we observe in the physical world can be described to arbitrary precision using computers; in particular, the QM behavior of the microtubules does not appear to be beyond the capacity of a TM to simulate. The mathematical support comes from the observation that we have no consistent theories of any notion of "computability" or "information processing" that is beyond the capacity of a TM (the various sorts of oracle computation are explicitly inconsistent by assumption as they make the counterfactual assumption that certain problems we know to be unsolvable have nevertheless been solved).

 

[rocketdodger]

I may be missing something here (again) but is westprog asking us to define an abstraction in terms of things which are at a lower level of abstraction? And is this not what Russell's Theory of Types was about?

Sorry, westprog, I continually feel we are talking at cross purposes and a whiteboard and bottle of whisky would help.

Yes, that seems to be what he wants.

 

[westprog]

See you do believe a dog is a rabbit!

They both have the EXACT same processes underlying their biology so OBVIOUSLY they are the same.

They don't have the EXACT same processes, because they have different DNA.

 

[westprog]

I find it indicative of honesty you bring to these discussions that you would repeat the statement



while purposefully excluding the rest of the reply made by drkitten:

I find it typical of the way you conduct these discussions that in spite of the fact that I've been jumping back and forth producing an enormous amount of material in response to four or five different posters, you accuse me of dishonesty because of a short delay in addressing one point among many. It's unfortunately typical.

 

[westprog]

OK, lets play. This is fun.

Remember my earlier definition of computation? Append it to include:

where "internal state" refers to a specific equivalence class of possible configurations of all PHYSICAL PARTICLES within the system and "external state" refers to a specific equivalence class of possible configurations of all PHYSICAL PARTICLES outside of the system, and "system" refers to a specific subset of PHYSICAL PARTICLES within the set of all PHYSICAL PARTICLES in the universe.

There. Is the definition "physical" yet?

It would help if you could give the entire thing. It seems fine, it just applies to pretty well any physical systems that I can imagine.

 

[westprog]

I may be missing something here (again) but is westprog asking us to define an abstraction in terms of things which are at a lower level of abstraction? And is this not what Russell's Theory of Types was about?

Sorry, westprog, I continually feel we are talking at cross purposes and a whiteboard and bottle of whisky would help.

I'm giving fairly pragmatic working definitions of what mathematics and physics are. I'm not trying to develop a Theory Of Types.

 

[westprog]

If it's a purely mathematical property, then one is implicitly accepting that mathematical properties can have physical consequences, and therefore a mathematical definition of computation such as Turing-equivalence has consequences that are part of the physical world and can be studied as such.

I find it hilarious that DrKitten is proposing that mathematics is really physics and at the same time insist that a mathematician can't be a physicist.

 

[westprog]

Not get the physics obviously wrong is a necessary but sufficient condition.

Yet you claim that he got the maths wrong as well.

 

[Wudang]

I'm giving fairly pragmatic working definitions of what mathematics and physics are. I'm not trying to develop a Theory Of Types.

No but from what I read you're trying to sidestep it. You're asking us to define abstracts in terms of the physical implementations or instantiations of the abstract in physical terms and then rejecting that because they aren't abstract enough.

 

[rocketdodger]

I find it typical of the way you conduct these discussions that in spite of the fact that I've been jumping back and forth producing an enormous amount of material in response to four or five different posters, you accuse me of dishonesty because of a short delay in addressing one point among many. It's unfortunately typical.

By using the term "delay" you imply that you responded to those statements or that you intend to respond to them.

You obviously have not responded to them yet. Do you intend to?

 

[rocketdodger]

It would help if you could give the entire thing. It seems fine, it just applies to pretty well any physical systems that I can imagine.

Computation occurs in a system S (a system S computes) when transitions between a set of external states E affect transitions between a set of internal states I such that |E| > |I|, where "internal state" refers to a specific equivalence class of possible configurations of all PHYSICAL PARTICLES within the system and "external state" refers to a specific equivalence class of possible configurations of all PHYSICAL PARTICLES outside of the system, and "system" refers to a specific subset of PHYSICAL PARTICLES within the set of all PHYSICAL PARTICLES in the universe.

 

[westprog]

However, there's a lot of support for the idea that the human brain is Turing equivalent, both physical and mathematical. The physical support comes from the observation that everything that we observe in the physical world can be described to arbitrary precision using computers; in particular, the QM behavior of the microtubules does not appear to be beyond the capacity of a TM to simulate. The mathematical support comes from the observation that we have no consistent theories of any notion of "computability" or "information processing" that is beyond the capacity of a TM (the various sorts of oracle computation are explicitly inconsistent by assumption as they make the counterfactual assumption that certain problems we know to be unsolvable have nevertheless been solved).

Yes, it's possible to simulate operations of the brain using computers. It's also possible to simulate the operations of volcanoes. Does that mean that volcanoes are Turing-equivalent?

 

[westprog]

Computation occurs in a system S (a system S computes) when transitions between a set of external states E affect transitions between a set of internal states I such that |E| > |I|, where "internal state" refers to a specific equivalence class of possible configurations of all PHYSICAL PARTICLES within the system and "external state" refers to a specific equivalence class of possible configurations of all PHYSICAL PARTICLES outside of the system, and "system" refers to a specific subset of PHYSICAL PARTICLES within the set of all PHYSICAL PARTICLES in the universe.

Yup, near as I can tell that applies to just about anything in the universe.

 

[westprog]

By using the term "delay" you imply that you responded to those statements or that you intend to respond to them.

You obviously have not responded to them yet. Do you intend to?

You seem to have some kind of idea that I'm accountable to you. I don't know how you got that idea, but you are very much mistaken.

 

[westprog]

No but from what I read you're trying to sidestep it. You're asking us to define abstracts in terms of the physical implementations or instantiations of the abstract in physical terms and then rejecting that because they aren't abstract enough.

No, I'm quite willing to accept an entirely abstract definition of computation, if it's well defined. However, I expect a physical definition to refer to physical properties in some way.

Which of the points of view do you find fit the bill so far? Rocketdodger has provided a definition of computing. I don't think it describes anything in particular. DrKitten, I think (it's difficult to keep track of everyone) says that it's possible in principle to define computing in a physical way, but in practice it's too cumbersome, so a functional definition is just fine. Pixy Misa says that it will all become clear to me if I just read some Wiki articles.

 

[Wudang]

No, I'm quite willing to accept an entirely abstract definition of computation, if it's well defined. However, I expect a physical definition to refer to physical properties in some way.

Which of the points of view do you find fit the bill so far? Rocketdodger has provided a definition of computing. I don't think it describes anything in particular. DrKitten, I think (it's difficult to keep track of everyone) says that it's possible in principle to define computing in a physical way, but in practice it's too cumbersome, so a functional definition is just fine. Pixy Misa says that it will all become clear to me if I just read some Wiki articles.

OK. How are you defining computing? Arithmetic? The "Principles of Operations" manual for z/OS describes how arithmetic operations work for the z/OS architecture. In physical terms there are special areas in a CPU called the PSW (where the current instruction is located) and there are registers which point to other areas of memory. These are treated as hexadecimal values which are added for example using AND, XOR etc gates in the hardware.
Any better?
How about "It's complicated"?

 

[westprog]

OK. How are you defining computing? Arithmetic? The "Principles of Operations" manual for z/OS describes how arithmetic operations work for the z/OS architecture. In physical terms there are special areas in a CPU called the PSW (where the current instruction is located) and there are registers which point to other areas of memory. These are treated as hexadecimal values which are added for example using AND, XOR etc gates in the hardware.
Any better?
How about "It's complicated"?

What we're looking for is a definition of computing (or computation, if you prefer) which is physical, and is defined for all possible forms of computing.

 

[Wudang]

What we're looking for is a definition of computing (or computation, if you prefer) which is physical, and is defined for all possible forms of computing.

Then I bow out as we're talking about maths and Dr Kitten has the bases covered.

 

[rocketdodger]

Yup, near as I can tell that applies to just about anything in the universe.

Um, no, it doesn't.

For example, a rock sitting in the sun. Are there always (at least) two external state transitions that map to a single state transition within the rock?

 

[westprog]

Um, no, it doesn't.

For example, a rock sitting in the sun. Are there any two external state transitions that map to a single state transition within the rock?

Of course there are, because we can define the state transitions to be anything we want.