On 09 Oct 2013, at 15:43, Craig Weinberg wrote:
On Wednesday, October 9, 2013 3:18:52 AM UTC-4, Bruno Marchal wrote:
On 08 Oct 2013, at 20:12, Craig Weinberg wrote:
On Tuesday, October 8, 2013 12:34:57 PM UTC-4, Bruno Marchal wrote:
On 08 Oct 2013, at 17:59, Craig Weinberg wrote:
Why isn't computationalism the consequence of quanta though?
Human computationalism does.
But I want the simplest conceptual theory, and integers are easier
to define than human integers.
I'm not sure how that relates to computationalism being something
other than quanta. Humans are easier to define to themselves than
integers. A baby can be themselves for years before counting to 10.
Phenomenologically? Yes.
Fundamentally? That does not follow. It took a long time before
discovering the Higgs-Englert-Brout Boson.
It doesn't have to follow, but it can be a clue. The Higgs is a
particular type of elementary phenomenon which is not accessible to
us directly. That would not be the case with Comp if we were in fact
using only computation. If our world was composed on every level by
computation alone,
Hmm.... It is not obvious, and not well known, but if comp is true,
then "our world" is not "made of" computations.
Our world is "only" an appearance in a multi-user arithmetical video
game or dream.
it wouldn't make much sense for people to have to learn to count
integers only after years of aesthetic saturation.
What can be computed other than quantities?
Quantities are easily computed by stopping machines, but most
machines does not stop, and when they introspect, the theory
explains why they get troubled by consciousness, qualia, etc. Those
qualia are not really computed, they are part of non computable
truth, but which still bear on machines or machine's perspective.
Then you still have an explanatory gap.
But that is a good point for comp, as it explains why there is a
gap, and it imposes on it a precise mathematical structure.
But there's nothing on the other side of the gap from the comp view.
You're still just finding a gap in comp that comp says is supposed
to be there and then presuming that the entire universe other than
comp must fit in there. If there is nothing within comp to
specifically indicate color or flavor or kinesthetic sensations, or
even the lines and shapes of geometry, then I don't see how comp can
claim to be a theory that relates to consciousness.
There is something in the comp theory which specifically indicate
qualia.
The gaps in the intensional nuances could very well do that.
How can anything which is non-computable bear on the computation of
an ideal machine?
That is the whole subject of en entire field: recursion theory, or
theoretical computer science.
Ok, so what is an example of something that specifically bridges a
kind of computation with something personal that comp claims to
produce?
That is technical, and you need to study AUDA. I would say that *all*
statements in X1* minus X1 produces that. No doubt many open problems
have to be solved to progress here.
But even if that fails, you have not produced an argument that it is
not possible.
What connects the qualia to the quanta, and why isn't the qualia
just quantitative summaries of quanta?
Qualia are not connected to quanta.
Then what is even the point of Comp? To me quanta = all that relates
to quantity and certain measurement. If they are not connected to
quanta then a machine that is made of quanta can't possibly produce
qualia that has no connection to it. That's no better than Descartes.
I realize that you have not yet really study comp. Physical Machine
are not made of quanta. Quanta appears only as first person plural
sharable qualia. They are observable pattern common to people
belonging to highly splitting or differentiating computations, most
plausibly the "linear computations" (like in QM).
Quanta are appearances in the qualia theory, and they are not
quantitative, they are lived at the first person plural views.
Quanta aren't quantitative?
They might be. The fact that they come from qualia does not prevent
that they have quantitative aspect.
If Arithmetic truth is full of non nameable things, what nameable
things does it also contain,
The numbers, the recursive properties, the recursively enumarable
properties, the Sigma_i truth, well a lot of things.
You have the recursive (the simplest in our comp setting), then the
recursively enumerable (the universal machines, notably), then a
whole hierarchy of non computable, but still nameable set of
numbers, or machine's properties,
You say they are nameable, but I don't believe you. It is not as if
a number would ever need to go by some other name. Why not refer to
it by its precise coordinate within Arithmetic Truth?
Because it is independent of the choice of the computational base,
like volume in geometry. If you can name something with fortran,
then you can name it with numbers, combinators, etc. Nameability is
"machine independent", like the modal logics G, G*, Z, etc;
What you are calling names should be made of binary numbers though.
I'm asking why binary numbers should ever need any non-binary, non-
digtial, non-quantitative names.
Your methodology cannot work. Even if I was not able to explain how
non-quantitive names appear, it is up to you, when saying that comp is
wrong, have to give he impossibility argument.
Then, in this case, I keep telling you why: the math shows why and how
those non quantitative relations develop, notably through an
intersection or conjunction between truth and self-reference.
then you got the non nameable properties, like true (for number
relations) but very plausibly, things like consciousness, persons,
etc.
Some of those non nameable things can still be studied by machines,
through assumptions, and approximations.
Above that you have the truth that you cannot even approximated, etc.
Arithmetical truth is big, *very* big.
Big, sure, but that's exactly why it needs no names at all.
It is worst than that. Many things cannot have a name.
what can they have?
Properties, relations, well, ... the nameable things are exceptional,
in fact. in arithmetic. Nameable sets of numbers are enumerable, non
nameable sets are non enumerable. Life and consciousness develops at
the frontier between the nameable, and the non nameable. Truth itself
conditions everything, yet is not nameable.
Each feature and meta-feature of Arithmetic truth can only be found
at its own address. What point would there be in adding a fictional
label on something that is pervasively and factually true?
In science it is not a matter of decision, but of verifiable facts.
That's what I am saying, what is the point of adding a fictional
name to a fact that is verifiable within Arithmetic truth?
?
and what or who is naming them?
The machines. (in the comp setting, despite the machines theology
does refer to higher non-machine entities capable of naming things.
That's the case for the first order logical G* (which I note
usually qG*, this one needs more than arithmetical truth, but it is
normal as it describes an intensional (modal) views by a sort of
God (Truth) about the machine. here the miracle is that its zero
order logical (propositional) part is decidable.
I don't think that names and machines are compatible in any way.
Programmers of machines might use names, but once compiled, all
high level terms are crushed into the digital sand that the machine
can digest. No trace of proprietary intent remains.
Not at all. The whole point is that such proprietary are invariant
for the high or low level implementations.
Then why would we have to compile high level programs into low level
machine code?
Because it is simpler for a human to talk in higher level terms, and
it is easier for the engineer to build a universal machine from low
level basic components.
Our brains have a low level (neuronal for example), and high level (as
it is apparent from its structure, not mentioning our introspection
ability).
The modal logics I am using does capture high level properties of low
level defined by addition+multiplication.
If quanta is Löbian qualia, why would it need any non-quantitative
names?
? (to fuzzy question, sorry, try to make this more clear perhaps).
You said earlier that quanta is Löbian qualia, and then you are
saying above that naming is one of the things that Löbians
(persons? machines?) do, so I am asking why don't they just use
quanta instead of non-quantitative qualia?
Because, by the logical gaps Z/Z*, X/X* etc. (inherited from G/G*)
it is shown that Löbian numbers/machines/person are confronted with
them. They can names some things, and they cannot name other things.
It is a theorem in their self-reference logics. It is "basic"
machine's theology or psychology.
It sounds like you're just saying "because that's how they work."
I say something like that, but I explain with computer science and
logic. Before Gödel we could think that machines are basically a
simple notion, amenable to complete theories. After Gödel, we realize
that above a threshold of complexity (the Turing universality
threshold) we know about nothing. And with comp, we can understand
why, as universal machines appears to be unknown to themselves,
uncontrollable by themselves, full of not predictible quantitative and
qualitative surprises.
This does not prove that comp is true, but it defeats argument based
on easy comparison between us and possible machines.
Bruno
http://iridia.ulb.ac.be/~marchal/
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/groups/opt_out.
