On 3/30/2012 2:48 PM, meekerdb wrote:
On 3/30/2012 4:08 AM, Stephen P. King wrote:
On 3/30/2012 3:08 AM, meekerdb wrote:
On 3/29/2012 10:23 PM, Stephen P. King wrote:
Take my favorite thought experiment. Suppose I design two Mars
Rovers and I want them to coordinate their movements in order to
round up Martian sheep. I can easily distribute the artificial
intelligence between the two of them, using data links so whatever
one sees the other sees (incidentally this, minus the AI, is what
combat aircraft software does now) and so there is a single top
level decision routine on top of local decision routines about
maneuvering around obstacles and managing internal states.
OK, I'll bite. That AI program is running on the combination of
the two pieces of hardware, this is no different than how Watson
<http://www.usatoday.com/tech/news/2011-01-14-ibm-jeopardy_N.htm>
runs on a much of connected servers. But think about it, does not
the data correlated to the sensors on one of the Rovers have to be
synchronized with the data from the other Rover so that their
manuevering can be controled. How exactly is the internal model of
this system built so that it can 'consider itself' as being both
exploring some pile of rocks east of Mt. Olympus while the other
Rover is taking a dirt sample in some crater 500 kilometers away.
It is not possible for two fixed points to exist on one compact and
closed manifold. You can only have one at a time. What you describe
is more like a Rover with a multiple personality disorder.
You're suffering from a failure of imagination. It can make a model
in which it considers itself has having to bodies at two different
places. It wouldn't even be hard to program. If you can experience
two different places and you can act in two different places you ARE
in two different places.
Brent
--
Hi Brent,
Could you give us a brief explanation of how this would work? I
will readily admit to a failure of imagination here.
Just as you have two eyes providing two 2D views of the world which
your brain merges into one 3D model, one can merge the data streams
from many different sensors located in disparate places to create a
single unified model. Military systems do this for battlefield
awareness. And this model can include a 'self'. The self is usually
located where the hardware is because evolution requires that the
hardware/self be protected.
Brent
Hi Brent,
Interesting. " ... merge ... streams ... to create a single unified
model...". How is this not consistent with my original claim above? My
claim is that a single unified model must obtain and that the notion of
1-p identity follows directly from this and thus if no unification can
obtain neither does a 1-p. What did you imagine that I was claiming?
What I have been trying to point out is that the "single unified model"
must be such that there are no logical or resource conflicts. I go
further to claim that this is an example of the SAT problem in applied
logic. It is known that SAT is an NP-Complete problem and thus the
computational complexity issues must be addressed.
I have been arguing that COMP neglects these complexity issues
completely and seemingly does so by eliminating the necessity to
consider the actual physical implementations of the computations,
preferring instead to postulate some "Platonic" realm where computations
have particular properties merely on the basis, it is argued, that the
truth valuations of Sigma_1 sentences is completely independent of our
knowledge of their content. I see this as equivalent to saying that
"since there exist strings of numbers that are solutions to NP-Complete
problems then we don't need to worry about actually having to find them"
and thus bypass problems like SAT. Instead we are offered an interesting
form of Occasionalism or Accidentalism to explain how it is that we
actually do experience a physical world.
Onward!
Stephen
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en.
