4 ms·

Consider the famous 100 year old paradox: this sentence is false. Watson won't be able to assign a truth value to this gramatically correct sentence, but would

Consider the famous 100 year old paradox: this sentence is false. Watson won't be able to assign a truth value to this gramatically correct sentence, but would it have the insight that it just stumbled upon a philosophical problem that's solved by narrowing the notion of truth it uses (Bertrand Russell). In other words, a machine may not be able to examine itself in the way that humans can when certain problems arise.

cousin_it 16y ago
Long ago, some people said computers would never beat humans at chess, and they were wrong. A human brain contains no magical components. It runs on physics. Whatever it does, can in principle be done by some other machine that runs on physics. I won't be surprised if someone finds a general way to replace metamathematical insight with clever search-and-pruning over a large but simply structured space, like it happened with chess.
- pnathan 16y ago
  That is an assumption regarding the human brain. Also, the limits of computation have been given an upper bound. If it's not computable by a Turing machine with a finite memory tape (or anything else that is mappable to a Turing machine with a finite memory tape), it's not computable by a physical computer. Steve Yegge started exploring the metaphysics of computer programs before he stopped blogging; it makes for interesting reading.
  - cousin_it 16y ago
    There's a reason why my argument never mentioned Turing computability :-) If a human brain uses quantum effects to get cognitive bonuses, you can build a quantum computer that does same. If the brain uses super-duper unified field theory, you can build a machine to take advantage of that, and so on. The only assumption I make about the human brain is that it doesn't run on magic juju that's available exclusively to humans by the orders of Cthulhu, and I think that assumption is quite warranted. (As far as I know, neurons can be approximated well enough by classical computers and you don't even need to go quantum, but that's another story.)
    - signa11 16y ago
      moon-is-a-harsh-mistress quote: Am not going to argue whether a machine can "really" be alive, "really" be self-aware. Is a virus self-aware? Nyet! How about oyster? I doubt it. A cat? Almost certainly. A human? Don't know about you tovarishch, but I am. Somewhere along evolutionary chain from macromolecule to human brain self-awareness crept in. Psychologists assert it happens automatically whenever a brain acquires certain very high number of associational paths. Can't see it matters whether paths are protein or platinum.
  - VladRussian 16y ago
    > If it's not computable by a Turing machine ... it's not computable by a physical computer. By a physical Turing equivalent computer. True. The breakthrough will come when we'll be able to overcome the damnation of step-by-step-ness (which is the primitive expression of causality).
Tichy 16y ago
I am a human and I also can not decide if that sentence is false or not. How does that make me better than a computer?
- Joakal 16y ago
  Logically, the question has no answer. I don't see how a computer won't say no answer in a given "Yes or No or N/A" too. It already happens for 1/0 answers.

cousin_it 16y ago

Long ago, some people said computers would never beat humans at chess, and they were wrong. A human brain contains no magical components. It runs on physics. Whatever it does, can in principle be done by some other machine that runs on physics. I won't be surprised if someone finds a general way to replace metamathematical insight with clever search-and-pruning over a large but simply structured space, like it happened with chess.

pnathan 16y ago
That is an assumption regarding the human brain. Also, the limits of computation have been given an upper bound. If it's not computable by a Turing machine with a finite memory tape (or anything else that is mappable to a Turing machine with a finite memory tape), it's not computable by a physical computer. Steve Yegge started exploring the metaphysics of computer programs before he stopped blogging; it makes for interesting reading.
- cousin_it 16y ago
  There's a reason why my argument never mentioned Turing computability :-) If a human brain uses quantum effects to get cognitive bonuses, you can build a quantum computer that does same. If the brain uses super-duper unified field theory, you can build a machine to take advantage of that, and so on. The only assumption I make about the human brain is that it doesn't run on magic juju that's available exclusively to humans by the orders of Cthulhu, and I think that assumption is quite warranted. (As far as I know, neurons can be approximated well enough by classical computers and you don't even need to go quantum, but that's another story.)
  - signa11 16y ago
    moon-is-a-harsh-mistress quote: Am not going to argue whether a machine can "really" be alive, "really" be self-aware. Is a virus self-aware? Nyet! How about oyster? I doubt it. A cat? Almost certainly. A human? Don't know about you tovarishch, but I am. Somewhere along evolutionary chain from macromolecule to human brain self-awareness crept in. Psychologists assert it happens automatically whenever a brain acquires certain very high number of associational paths. Can't see it matters whether paths are protein or platinum.
- VladRussian 16y ago
  > If it's not computable by a Turing machine ... it's not computable by a physical computer. By a physical Turing equivalent computer. True. The breakthrough will come when we'll be able to overcome the damnation of step-by-step-ness (which is the primitive expression of causality).

Tichy 16y ago

I am a human and I also can not decide if that sentence is false or not. How does that make me better than a computer?

Joakal 16y ago
Logically, the question has no answer. I don't see how a computer won't say no answer in a given "Yes or No or N/A" too. It already happens for 1/0 answers.