>
> Thanks, Colin,
> I feel we also agree in your last sentence statement, however I could
not
> decide whether "abstraction" is reductionist model forming or a
> generalization into wider horizons? Patterns - I feel - are IMO
definitely reductive.
Abstraction I would characterise as a mapping into a representational
domain. As to the level of reduction, that would depend on the domain of
symbols and their mapping to the mapped domain. The questions to ask
yourself are:
a) Who decides what the lowest level domain is to be?
b) What do you lose when you choose?
Let's look at abstracting a whole human:
a-1) Say we 100% abstract a human down to a representation of cells. Cells
would be the base level descriptive domain. Organs would be data patterns
that the cells express under the rules of the abstraction. And so on. You
are not letting the natural rules run. You are merely moving symbols
around. No matter how powerful the computer and how detailed (lowest level
domain) the abstraction is just the computer's representation of the
symbols being moved around.
a) If you build a squillion little computers, each to act 'as-if' they
were the, say, the cell level of the abstraction with a little physical
interface that meant it was just like a cell from the outside, then you
have reinstated some level of the natural world's involvement...the
resulting human may be indistinguiishable from a human. Organs are an
emergent property of these collaborating 'Turing Cells'.
b) Then again, if you reduced the abstraction level to build tiny
computers that become substitute molecules, so to all intents they looked
like molecules...the human would look the same. Cells are an emergent
property of collaborations of these 'Turing Molecules'. (please ignore the
need for fluids and food etc in this body for the moment!)
If you inspected the human a-1) at the molecular level all you see is a
computer playing with patterns depending on the chosen abstraction domain.
If you inspected human a) at the molecular level you would see only the
computer that runs cells, but the cells would look normal. There are no
human molecules here, only the molecules of the computers inside the
cells.
If you inspected human b) at the molecular level you would see what
appears to be real molecules. The cells would look normal. However, look
for atoms and you won't find any.
Q. What is it like to be human a-1) cf a) cf b) and how well does each
human operate cognitively?
You could extend the argument to simulated Turing-quarks and
Turing-leptons... and so on... at some point the human would acquire
consciousness. What level of Turing-granularity is that?
My answer to this would be probably waaaaaay down deep below where the
matter and the space differentiate their behaviour. We have no
justification that any one level of organisation is an end-point.
>
> that scale-game (40-50 orders of m. down) seems to me valid within the
physical explanatory equationalized circumstances - so I scrutinize it
(accept it within physics-thinking). It does not refer to 'time'
(whichever
> you prefer). I had the notion that there 'is' only "change" ie. movement
and
> space is a time-coordinate of it, while time is a space-coordinate of
it,
Change as a structural primitive is quite workable. Imagine being human
shaped water in one place in a waterfall... i.e. regular structure within
change. At any instant there is a human, but the water is flowing, so the
componentry of the human is dynamically refreshed. Think of "humanity".
Humanity survives where all the humans in it don't. Same thing at all
scales.
An infinity of potential collaborations of that one tiny change primitive
that have a net value of "1 change primitive" can be substituted for any
other change primitive. This recursiveness is the basis of a
calculus/logic.
In this system time results merely from the state of the collaboration
undergoing a transition as the change primitive does what it does (eg
changes from state A to B then back to A). There's not such thing as time
in this structure. If the state changes happen at a regular enough rate
then equations with a "t" in it are possible as descriptors. The universe
acts 'as-if' there was time. If you are made of a pile of these changes
then, if there was an
observation faculty, all you would see is the collaboration evolving
according to the rules of the structural primitives. You would need to
"see" only the structural regularity, not the change primitives.
In the waterfall metaphor, two humans as regularity in this waterfall
would not see any water. They would see only each other and the space in
between. This is nature of
In this structural domain these things are really simple.
Also: If you take a slice _across_ this structure around the level of
atoms, photons etc and devise mathematical descriptions for the behaviour
of identified structures you get quantum mechanics. QM says absolutely
nothing about 'what it is that is behaving quantum mechanically.
There is nothing there but the one single change primitive. Call it X.
Everything is merely organisations of X. Stare at a human and all you are
staring at is X. There are no 'things' atoms. The atoms are all X behaving
like atoms.
That is where I would take you idea of change.
You can relate this directly to the 'Turing-granulation' discussion above
easily enough.
> as
> long as we think in terrestrially (not even of THIS universe) formed
explanations of those figments we conclude upon the latestly primitive
instrumental observations in our reductionist science domain.
> Matter and its 'behavior' is similarly 'concluded' to reflect the
'personal'
> experiencing of the unknown effects.
>
> I am deterred by the semantic direction of 'computing'. If it is Bruno's
manipulation of ordinary numbers, I feel OK, but then I feel domains of
incompetence. Your "as-if" changes that and I felt lost. Why use a word
with 'other' meaning 'as - if'? It is a cheap excuse that we have no
better
> one <G>.
>
> Sorry for just "multiplying the words" in this exchange.
> John M
>
I'm having trouble following your issue. My background is one of a huge
involvement in software and computers from the very highest to the very
lowest levels. There's nothing like building basic aritmetic out of logic
gates to get you understanding what is going on!
I find it easy to see that none of the mathematical domains we define,
including the natural numbers, are ever actually reified in computing.
Take a variable X that is supposed to represent a quanity. There is a
memory location where some electrons are thrown in pattern of electron
buckets that by our agreed convention encodes the quantity as a binary
representation. We then have another pointer called X that says where the
buckets are.
We write progams to manipulate the electron buckets. The pattern in the
electron buckets acts 'as-if' or stands for a real platonic 'thing' that
is intrinscally a quanitiy X. I find it impossible to imagine this
platonic object that is intrinsically "X_quantityness". What I do know for
sure is that nowhere in the computer is this platonic "thing". All there
is are electron buckets going through a pattern contrived to behave in the
same way as the platonic object would if it existed.
Not sure if that helps... but the idea of abstractions being 'as-if'
representations seems a useful way of grounding ourselves.. to beware that
the so-called reality we are part of is implemented in a domain unlike any
we currently explore, and that it is validly considered to be computation,
it's just that the computation is very different.
Not sure if that helps... but... well it was fun to write!
cheers
colin hales
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list.domain.name.hidden
To unsubscribe from this group, send email to everything-list-unsubscribe.domain.name.hidden
For more options, visit this group at
http://groups.google.com/group/everything-list
-~----------~----~----~----~------~----~------~--~---
Received on Fri Jul 28 2006 - 22:18:58 PDT