Re: Turing Machines and Physical Computation

From: Bill Modlin (wdmodlin_at_yahoo.com)
Date: 11/30/04 

Date: Tue, 30 Nov 2004 16:32:25 -0500

"patty" <pattyNO@SPAMicyberspace.net> wrote in message
news:VQ4rd.590978$mD.327332@attbi_s02...
> Eray Ozkural exa wrote:
>
> > Neil W Rickert <rickert+nn@cs.niu.edu> wrote in message
news:<coe8th$tdu$1@usenet.cso.niu.edu>...
> >
> >>>The putative difficulty in writing effective software is a more common
> >>>and mundane version of the same thing, IMHO.
> >>
> >>That's actually a quite different problem.
> >>
> >>Computation is the manipulation of representations.
> >
> >
> > I am not sure if that's a good description.
> >
> > This formal symbol manipulation idea got some otherwise ambitious
> > philosophers like Brian Cantwell Smith and gang quite confused.
> >
> > Computers can work on representations, that's true. But it is not
> > necessary that what is being manipulated is "representation".
> >
>
> Can you give us an example of where a computer is *not* manipulating a
> representation ?
>
> patty

Patty, weren't you the one who suggested that representation is only
meaningful in the sense of representing something to an interpreting
observer? The machine goes through a sequence of changes in state.
Normally the people who program it think of some elements of those states as
representing something... numbers, characters, cars, hurricanes, chess
pieces, whatever. But there may be ambiguity about what a particular
element represents, depending on ones point of view. There is nothing about
for example the state of a memory location that makes it inherently a
representation of any particular type of thing. And if a computer were to
be executing some arbitrary (random, or of unknown origin) sequence of
operations, which was not a representation of anything particular to any
existing observer, I'm not sure in what sense you could say that it was
manipulating a representation. It is just going through a sequence of
causally dependent material state changes.

Of course, one could take the position that such a sequence is no longer a
computation. One could say that "computing" *means* manipulating
representations. Which might be a good thing, since it explains why we are
uncomfortable saying that a projectile or a planet "computes" its
trajectory. We only ascribe "computation" to physical processes that we
take as representing other things. And with "representation" dependendent
on an interpreting observer, all arguments about whether there really *are*
computations and representations in the brain or anything else become
moot... it all depends on whether you choose to so interpret them.

Bill

Relevant Pages

  • Re: Turing Machines and Physical Computation
    ... weren't you the one who suggested that representation is only ... The machine goes through a sequence of changes in state. ... depending on ones point of view. ... One could say that "computing" *means* manipulating ...
    (comp.theory)
  • Re: Impossible Database Design?
    ... use a discrete representation of a subset of the reals, ... though the reals are continuous. ... the someone manipulating you for his own self-promotion. ... but into granting him an intellectual high ground. ...
    (comp.databases.theory)
  • Re: Epistemology 201: The Science of Science
    ... representation of the number represented, add, transform the result into ... algorithms for maniuplating strings. ... that most people find arithmetic so hard is that it is the manipulating ...
    (sci.math)
  • Re: Epistemology 201: The Science of Science
    ... representation of the number represented, add, transform the result into ... algorithms for maniuplating strings. ... that most people find arithmetic so hard is that it is the manipulating ...
    (sci.cognitive)
  • Re: Epistemology 201: The Science of Science
    ... representation of the number represented, add, transform the result into ... algorithms for maniuplating strings. ... that most people find arithmetic so hard is that it is the manipulating ...
    (sci.physics)