Hallo!

On Sun, 1 Aug 2010 20:14:58 +0100, R A Brown wrote:

>  Jörg Rhiemeier wrote:
>  >  Hallo!
>  >
>  >  On Sat, 31 Jul 2010 20:44:36 +0100, R A Brown wrote:
>  [snip]
>  >>
>  >>   One thing's certain - the human brain ain't a von Neumann
>  >>   machine  ;)
>  >
>  >  This is indeed certain.  I would rather guess that it is a quantum
>  >  information system.
>
>  Which IIRC is what Roger Penrose argues for in "The
>  Emperor's New Mind."

I found this idea in a book written by a physicist and a psychologist,
_Der kreative Kosmos_ by Thomas and Brigitte Görnitz.  But it is
obvious.  Classical systems are deterministic (even if it can become
practically impossible to predict their behaviour) and therefore
cannot have free will.  Quantum systems are not deterministic.

This would also ruin all attempts to "upload" the human mind onto a
computer, as many transhumanists dream of.  Apart from the practical
problem that it would have to be done in zero time because the mind
is a highly dynamic system, you first need a quantum computer; any
attempt at uploading a human mind onto a classical computer would
necessarily cause its wave function to collapse, thereby killing
the subject.  Then, there is the no-cloning theorem.  While classical
information can, as we all know, be copied as often as one wants,
this does not work with quantum information.  It is possible to
transfer quantum information from one system to another - but in
the process, the original is always destroyed.

In my planned science fiction novel _Schönster aller Planeten_
about a transhumanist colony on Mars, those transhumanists try to
upload minds several times - and always kill the subject and only
obtain a useless mess of classical information rather than an
uploaded mind.

>  >  I think that is the only way how human beings
>  >  could have free will - which they evidently do, despite some
>  >  philosophers claiming the opposite.
>
>  Yes, I think I have freely chosen to send this reply and
>  that my writing&  sending it has not been predetermined by
>  the inexorable cosmic forces.
>
>  But to return to the topic (unless it's been pre-ordained
>  otherwise    ;)

Sure.

>  ============================================
>
>  Lee wrote:
>   >  I think we are confusing parsing with understanding, and
>   >  human languages with programming languages.
>
>  I think Lee is right on both points.
>
>  We have noted before on this list that calling programming
>  code 'language' can be misleading; if we're not careful we
>  slip into the same carelessness as those who talk about
>  computers 'understanding' something or 'thinking' something
>  and really seem to assume that computers actually do think
>  or understand! They do nothing of sort.

Indeed not.  The use of the word "language" in expressions such
as _programming language_ is only a metaphor, and it is indeed
misleading in certain ways.  Computers do not think, and thus
never understand *anything*; they just mindlessly follow orders.

>  A programming 'language' is a very different beast from
>  human language and, as I say, we have gone over more than
>  once before on this list.
>
>  Also I think Lee does well to remind us that, as far as
>  humans are concerned, understanding what we hear or read is
>  quite a different process from parsing.  I remember we had
>  to parse English (and occasionally Latin or French)
>  sentences when I was at school.  Basically what we did was
>  to go through the sentence word by word, identifying what
>  part of speech it was, its morphology (except we didn't call
>  it that then) and how the word related to the rest of the
>  sentence.

Which is not the usual way of processing language.  It is an
artificial procedure which probably has very little to do with
the way how we understand language.

>  But it has already been observed in this thread that
>  something rather different is going on as we read or hear
>  language.
>
>  Whether it is easier, say, to _parse_ typical Welsh
>  sentences than it is to parse English ones, I don't know.
>  AFAIK there has been no study to test the relative ease or
>  difficulty of parsing SVO, SOV and VSO languages.  Besides,
>  other factors than mere word order will .also play a part.
>  e.g. in some languages morphology will make things clearer
>  than in others.

Right.  And as I have observed earlier in this thread, the simplest
languages to *parse* are RPN-like, stack-based languages such as Fith.
They require as little parsing overhead as possible.  No parse trees,
no temporary storage for items read but not yet assigned, just a stack
- the simplest of all compound data structures.  Yet, stack-based
languages are very difficult for a human being to *understand*: Fith,
with all that center embedding and other syntactic tricks its grammar
allows, just cannot be processed by the human brain in real time; also,
programming in FORTH and its ilk is more difficult than programming in
a more "human-like" programming language.

--
... brought to you by the Weeping Elf
http://www.joerg-rhiemeier.de/Conlang/index.html