2009/1/30 Bruno Marchal <marchal.domain.name.hidden>
>
> On 29 Jan 2009, at 20:42, Quentin Anciaux wrote:
>
>
>
>>
>> Why would the movie graph rule out a notion of *computational*
>> supervenience. We can keep comp and abandon materialism. We can still say
>> yes to the digitalist doctor, by betting on our more probable
>> relative computational histories.
>>
>
> Because here there was no "physical" graph at the start... it's just a
> program and we replace various subpart of it which instead of computing make
> a lookup. Now frow what you are saying I should understand that the program
> is a "description" of a computation, not a computation
>
>
>
> Not really.
> A program is a finite piece of information capable of making computing some
> universal machine. The computation is the activity (abstract or physical,
> etc.) of that universal machine.
>
Ok then for the particular run I describe, the two programs (the original
and the one modified by stub subpart) have the same states... So for this
particular run, we should still accept that the stub modified program is
conscious while the run is performed (either physicaly or just by the mere
existence of it's trace in platonia). Is consciousness supervening on the
state/trace of the computation (which makes that there is an infinity of
computations "outputing" this trace, and in our case original program and
stub program does the same) or on the trace+the relation between states
(which would differentiate the two programs, hence by what supervening mean
the consciousness generated by them) ?
Another thing is that the meaning (3rd pov) of the computation itself/states
is dependant on an observer... if you can see the stack of number of a
computation without the program and the current instruction you have no way
to tell it isn't just noise.
>
> and "consciousness" supervene on computation but not description of
> computation ? Then I'm lost about what is a computation ?
>
>
>
>
> Here we met a typical difficulty, a bit like showing the moon to a cat. The
> cat looks only to the finger. The difficulty is illustrated by famous
> paintings of Magritte: "ceci n'est pas une pipe" (a drawing of a pipe, with
> the mention: this is not a pipe).
>
> Except that this difficulty is more severe in our context.
>
>
>
>
> I would thing a computation is the act of "running" the program, execute
> each step and modifying the internal state.
>
>
>
> Well it is exactly that, except that you are ambiguous if the act has to be
> physically implemented or could be realized into purely numeric relations.
>
> But it is exactly what you say: so a computation will usually be described
> by a logical sequence of "states": A, B, C, D, ...
>
> And what I am saying is that comp makes consciousness supervening on A, B,
> C, D, ... (a computation), but that consciousness does not supervene on a
> description (in some language) of the computation "A, B, C, D, ...".
> It is really the difference between A, B, C, D, ..., and "A; B, C, D,
> ..."
>
> Even in the abstract we have to distinguish something like a number, a
> machine, a theory, a proof, a computation, from their abstract (but coded)
> descriptions.
>
Ok
>
> Later I will probably describe a computation by a number (PA handles only
> numbers). Consciousness does not supervene on such a number. It will appear
> that if PA (or RA) proves the existence of that numbers together with an
> account that it represents
> a computation, such a proof will be truly (but not necessarily provably)
> equivalent to a computation.
>
> Of course this is a bit subtle, and for being clear we have to be technical
> and provide definitions. I will say more in the seven step.
> It is subtle because computations themselves uses descriptions of things,
> so confusions can grow, and one confusion on one level extends on all
> levels!
>
> Your definition of computation is correct. In platonia, a computation is
> NOT automatically equivalent to a *description* of a computation. Why would
> the finger and the moon collapse in platonia?
>
>
Ok, I still have to think at it.
.
>
>
>
>
> I'm talking about a modified version of the movie graph where instead of
> starting with a conscious physical gates machine, I start with a conscious
> program and transform it accordingly (broken gates + projection of the film
> in the movie graph, in my case, stub subpart which do a lookup instead of
> computing the value)
>
>
>>
>> The absurdity with the movie graph is that it shows that associating
>> consciousness to the *physical* implementation leads us to attribute
>> consciousness to a description of a computation, and that is ridiculous.
>> (Well at least, once you understand what is a (mathematical) computation,
>> I think many are confused here, I will come back on this).
>> Computations and descriptions of computation are related, but differs and
>> are not of the same type, nor level. It is hard to be clearer without going
>> through computer science. Probably more in my next post to Kim on the
>> Seventh Step.
>>
>> I am not really sure I get your "ruling out all kinds of supervenience".
>>
>
> Because from what I understand (which is surely wrong) it's absurd at
> all... (supervenience of consciousness) :
>
>
>
>
> But if you say yes to the doctor, you do accept that your (local and
> relative) consciousness will or should supervenes on something. You would
> not say yes to a doctor who proposes the simple brain ablation experience.
>
I do accept if there is a program that can generates me for all practical
purposes and if there is then I accept to be run on anything that can run
such a program, including the whole timeless running in platonia. But that's
the whole point, I must believe I'm turing emulable (which I believe)... But
I think we should be able to prove we are without needing to believe it is
true...
>
>
>
>
> from wikipedia:
> A supervenes on a set of properties B, if and only if any two objects x and
> y which share all properties in B (are "B-indiscernible<http://en.wikipedia.org/wiki/Indiscernible>")
> must also share all properties in A (are "A-indiscernible").
>
>
>
>
> Very good definition. So a naturalist will say that some consciousness (A)
> supervenes on some dynamic of a brain (B), if any dynamics x and y
> sharing the dynamic of the brain (B), will share the same consciousness
> (A).
> Now if we assume comp, those dynamics are turing emulable, and this, by
> UDA, makes consciousness (of one person) supervening on "sheaf of possible
> digital dynamic (computations) . It imposes constraints in the space of
> possible dynamic and constraints about what self-referential entities can
> bet about their most probable digital dynamics.
>
> Tell me if this has helped. Your questioning is very relevant.
>
> Bruno
>
>
> http://iridia.ulb.ac.be/~marchal/ <http://iridia.ulb.ac.be/%7Emarchal/>
>
>
It helps... but some days I think I understand the whole thing and other
days I'm no more confident :), I have a little bootstrapping problem... like
something is missing in platonia.
Regards,
Quentin
>
>
>
> >
>
--
All those moments will be lost in time, like tears in rain.
--~--~---------~--~----~------------~-------~--~----~
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?hl=en
-~----------~----~----~----~------~----~------~--~---
Received on Sat Jan 31 2009 - 06:47:37 PST