Re: Re: Re: Where's the fixed identity in turing machines and comp ?

2012-11-12 Thread Roger Clough
ROGER: On Sat, Nov 10, 2012 at 05:55:03AM -0500, Roger Clough wrote: Hi Russell Standish No, rational beings have to decide which truths they need to apply to what and how to apply them. These are all relational acts, which require choice, hence intelligence. RUSS: I will insist that

Re: Re: Where's the fixed identity in turing machines and comp ?

2012-11-12 Thread Roger Clough
Hi Russell Standish Reason is what allows us to exist in the face of desire and danger. Roger Clough, rclo...@verizon.net 11/12/2012 Forever is a long time, especially near the end. -Woody Allen - Receiving the following content - From: Russell Standish Receiver:

Re: Where's the fixed identity in turing machines and comp ?

2012-11-12 Thread Stephen P. King
On 11/12/2012 2:53 AM, Quentin Anciaux wrote: 2012/11/11 Stephen P. King stephe...@charter.net mailto:stephe...@charter.net On 11/11/2012 11:15 AM, Bruno Marchal wrote: On 10 Nov 2012, at 17:44, meekerdb wrote: On 11/9/2012 3:26 PM, Stephen P. King wrote: It seems

Re: Where's the fixed identity in turing machines and comp ?

2012-11-12 Thread Bruno Marchal
On 11 Nov 2012, at 23:43, Stephen P. King wrote: On 11/11/2012 11:15 AM, Bruno Marchal wrote: On 10 Nov 2012, at 17:44, meekerdb wrote: On 11/9/2012 3:26 PM, Stephen P. King wrote: It seems to me that we automatically get a 'fixed identity' when we consider each observer's 1p to be

Re: Where's the fixed identity in turing machines and comp ?

2012-11-11 Thread Bruno Marchal
On 10 Nov 2012, at 17:44, meekerdb wrote: On 11/9/2012 3:26 PM, Stephen P. King wrote: It seems to me that we automatically get a 'fixed identity' when we consider each observer's 1p to be defined by a bundle or sheaf of an infinite number of computations. The chooser of A and of B

Re: Where's the fixed identity in turing machines and comp ?

2012-11-11 Thread Stephen P. King
On 11/11/2012 12:53 AM, Russell Standish wrote: On Sat, Nov 10, 2012 at 06:44:36PM -0800, meekerdb wrote: On 11/10/2012 5:37 PM, Russell Standish wrote: Only for some extended, loose definition of rational. The non-deterministic choices themselves are not rationally determined. Of course not

Re: Where's the fixed identity in turing machines and comp ?

2012-11-11 Thread Stephen P. King
On 11/11/2012 11:15 AM, Bruno Marchal wrote: On 10 Nov 2012, at 17:44, meekerdb wrote: On 11/9/2012 3:26 PM, Stephen P. King wrote: It seems to me that we automatically get a 'fixed identity' when we consider each observer's 1p to be defined by a bundle or sheaf of an infinite number of

Re: Where's the fixed identity in turing machines and comp ?

2012-11-11 Thread meekerdb
On 11/11/2012 12:59 PM, Stephen P. King wrote: I like this definition that was posted in that forum: *An agent is rational if he/she does whatever the modeler (i.e. economist) would do in his/her position.* The problem is with does. Flipping a coin and doing X if heads and Y if tails can

Re: Where's the fixed identity in turing machines and comp ?

2012-11-11 Thread Stephen P. King
On 11/12/2012 12:15 AM, meekerdb wrote: On 11/11/2012 12:59 PM, Stephen P. King wrote: I like this definition that was posted in that forum: *An agent is rational if he/she does whatever the modeler (i.e. economist) would do in his/her position.* The problem is with does. Flipping a coin

Re: Where's the fixed identity in turing machines and comp ?

2012-11-11 Thread meekerdb
On 11/11/2012 10:13 PM, Stephen P. King wrote: n 11/12/2012 12:15 AM, meekerdb wrote: On 11/11/2012 12:59 PM, Stephen P. King wrote: I like this definition that was posted in that forum: *An agent is rational if he/she does whatever the modeler (i.e. economist) would do in his/her

Re: Where's the fixed identity in turing machines and comp ?

2012-11-11 Thread Quentin Anciaux
2012/11/11 Stephen P. King stephe...@charter.net On 11/11/2012 11:15 AM, Bruno Marchal wrote: On 10 Nov 2012, at 17:44, meekerdb wrote: On 11/9/2012 3:26 PM, Stephen P. King wrote: It seems to me that we automatically get a 'fixed identity' when we consider each observer's 1p to

Re: Re: Where's the fixed identity in turing machines and comp ?

2012-11-10 Thread Roger Clough
Hi Russell Standish No, rational beings have to decide which truths they need to apply to what and how to apply them. These are all relational acts, which require choice, hence intelligence. Roger Clough, rclo...@verizon.net 11/10/2012 Forever is a long time, especially near the end. -Woody

Re: Where's the fixed identity in turing machines and comp ?

2012-11-10 Thread meekerdb
On 11/9/2012 3:26 PM, Stephen P. King wrote: It seems to me that we automatically get a 'fixed identity' when we consider each observer's 1p to be defined by a bundle or sheaf of an infinite number of computations. The chooser of A and of B is one and the same if and only if the

Re: Where's the fixed identity in turing machines and comp ?

2012-11-10 Thread Stephen P. King
On 11/10/2012 11:44 AM, meekerdb wrote: On 11/9/2012 3:26 PM, Stephen P. King wrote: It seems to me that we automatically get a 'fixed identity' when we consider each observer's 1p to be defined by a bundle or sheaf of an infinite number of computations. The chooser of A and of B is one and

Re: Where's the fixed identity in turing machines and comp ?

2012-11-10 Thread meekerdb
On 11/10/2012 9:58 AM, Stephen P. King wrote: On 11/10/2012 11:44 AM, meekerdb wrote: On 11/9/2012 3:26 PM, Stephen P. King wrote: It seems to me that we automatically get a 'fixed identity' when we consider each observer's 1p to be defined by a bundle or sheaf of an infinite number of

Re: Re: Where's the fixed identity in turing machines and comp ?

2012-11-10 Thread Russell Standish
On Sat, Nov 10, 2012 at 05:55:03AM -0500, Roger Clough wrote: Hi Russell Standish No, rational beings have to decide which truths they need to apply to what and how to apply them. These are all relational acts, which require choice, hence intelligence. I will insist that this is

Re: Where's the fixed identity in turing machines and comp ?

2012-11-10 Thread meekerdb
On 11/10/2012 2:54 PM, Russell Standish wrote: On Sat, Nov 10, 2012 at 05:55:03AM -0500, Roger Clough wrote: Hi Russell Standish No, rational beings have to decide which truths they need to apply to what and how to apply them. These are all relational acts, which require choice, hence

Re: Where's the fixed identity in turing machines and comp ?

2012-11-10 Thread Russell Standish
On Sat, Nov 10, 2012 at 03:27:47PM -0800, meekerdb wrote: But the definition [of rationality] seems overly restrictive. It's well known that in competitive games the best strategy may random in some way. So I don't see how you can arbitrarily rule out random choices as 'irrational'

Re: Where's the fixed identity in turing machines and comp ?

2012-11-10 Thread meekerdb
On 11/10/2012 3:56 PM, Russell Standish wrote: On Sat, Nov 10, 2012 at 03:27:47PM -0800, meekerdb wrote: But the definition [of rationality] seems overly restrictive. It's well known that in competitive games the best strategy may random in some way. So I don't see how you can arbitrarily

Re: Where's the fixed identity in turing machines and comp ?

2012-11-10 Thread Russell Standish
On Sat, Nov 10, 2012 at 04:37:55PM -0800, meekerdb wrote: On 11/10/2012 3:56 PM, Russell Standish wrote: But if it is rational to be irrational, is it possible to be rational any more? No, but you're making a conundrum out of it. The point is that it's rational to be non-deterministic.

Re: Where's the fixed identity in turing machines and comp ?

2012-11-10 Thread meekerdb
On 11/10/2012 5:37 PM, Russell Standish wrote: On Sat, Nov 10, 2012 at 04:37:55PM -0800, meekerdb wrote: On 11/10/2012 3:56 PM, Russell Standish wrote: But if it is rational to be irrational, is it possible to be rational any more? No, but you're making a conundrum out of it. The point is

Re: Where's the fixed identity in turing machines and comp ?

2012-11-10 Thread Russell Standish
On Sat, Nov 10, 2012 at 06:44:36PM -0800, meekerdb wrote: On 11/10/2012 5:37 PM, Russell Standish wrote: Only for some extended, loose definition of rational. The non-deterministic choices themselves are not rationally determined. Of course not by your definition of rational for in that case

Re: Where's the fixed identity in turing machines and comp ?

2012-11-10 Thread meekerdb
On 11/10/2012 9:53 PM, Russell Standish wrote: Correct. A stochastic decision is obviously not reasoned, so the decision itself cannot be rational. But that wasn't the original assertion. You said that a rational person could, necessarily, on chose one action. So if the rational decision is

Re: Where's the fixed identity in turing machines and comp ?

2012-11-09 Thread Bruno Marchal
On 09 Nov 2012, at 12:01, Roger Clough wrote: Hi Bruno In my discussions of intelligence, I define intelligence as the ability to (fairly freely) make one's own choices. OK. That implies through the word own that there is a fixed identity, the chooser OK. A picture of this might

Re: Where's the fixed identity in turing machines and comp ?

2012-11-09 Thread Russell Standish
On Fri, Nov 09, 2012 at 06:01:04AM -0500, Roger Clough wrote: Hi Bruno In my discussions of intelligence, I define intelligence as the ability to (fairly freely) make one's own choices. A bit of an odd definition, don't you think? A purely rational being does not have a free choice, they