From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!wupost!uunet!mcsun!uknet!newcastle.ac.uk!turing!ncmh Thu Jan  9 10:33:41 EST 1992
Article 2513 of comp.ai.philosophy:
Xref: newshub.ccs.yorku.ca comp.ai.philosophy:2513 sci.philosophy.tech:1729
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!wupost!uunet!mcsun!uknet!newcastle.ac.uk!turing!ncmh
>From: Chris.Holt@newcastle.ac.uk (Chris Holt)
Newsgroups: comp.ai.philosophy,sci.philosophy.tech
Subject: Re: Causes and Reasons
Message-ID: <1992Jan6.145723.1189@newcastle.ac.uk>
Date: 6 Jan 92 14:57:23 GMT
References: <1991Dec25.042628.18737@bronze.ucs.indiana.edu>   <1991Dec25.015221.6911@husc3.harvard.edu> <1991Dec28.221923.17443@bronze.ucs.indiana.edu> <1992Jan6.001554.7136@husc3.harvard.edu>
Sender: news@newcastle.ac.uk (NEWS system)
Organization: University of Newcastle upon Tyne, UK, NE1 7RU
Lines: 32
Nntp-Posting-Host: turing

Just a couple of quick asides in this thread.

zeleny@zariski.harvard.edu (Mikhail Zeleny) writes:
>                                        ...  The interesting claim is that
>the mind-brain relation is analogous to that of software to hardware; once
>you consider certain common properties of programs (e.g. finitude), you
>will surely agree that this is indeed an epistemic claim, albeit not one
>implying that the program in question be discoverable by human means.

(i) The mind-brain relation uses two distinct vocabularies to describe
different levels of abstraction; the relationship between these
levels is not understood.  The soft/hardware relation connects levels
of abstraction in precise ways, but there are many such levels, and
the choice of boundary between software and hardware can be varied:
if you want to implement an operation in hardware, you build it; if
you want to implement an operation in software, you simulate it.
So the strength of the analogy is unclear.

(ii) Finitude is not a straightforward property of a program.
Programs can create arbitrarily large data structures, and these
data structures may be programs, that the creator can execute.
Without finite temporal bounds, there need be no bounds on the
size of the program being executed; it is potentially infinite
(though not completed).  The original finite program may be
thought of as a program schema.  In a similar way, an axiom
schema (e.g. the Axiom Schema (V) of Subsets) in ZF is a finite
representation of an infinite number of axioms.

-----------------------------------------------------------------------------
 Chris.Holt@newcastle.ac.uk      Computing Lab, U of Newcastle upon Tyne, UK
-----------------------------------------------------------------------------
        "Between the motion / And the act / Falls the interface."


