On 15 Sep 2011, at 20:25, meekerdb wrote:
On 9/15/2011 10:34 AM, Bruno Marchal wrote:
Hi Evgenii,
On 13 Sep 2011, at 21:45, Evgenii Rudnyi wrote:
Bruno,
As I have already mentioned, I am not that far to follow your
theorem. I will do it presumably the next year.
Take your time. I am at the step 6 on the dot forum, where things
are done slowly, deeply and in a nice atmosphere :)
I have been working for the last ten year with engineers and my
consideration is so far at the engineering level.
All my work has been possible thanks to engineers, not scientist
nor philosopher who are still too much in the "the boss is right"
type of philosophy. To be sure there are some exceptions. But
usually engineers have a much better common sense and lucidity,
than scientist who seems to want to believe religiously in their
theories.
After all, if we know something, we should be able to employ it in
practice. And if this does not work in practice, then how do we
know that our knowledge is correct.
Working in practice does not mean truth.
Said that, I understand the importance of theory and appreciate
the work of theoreticians. After all, if we say A, then we must
say B as well. Hence it is on my list to follow your theorem (but
not right now).
No problem.
At present, I am just trying to figure out our beliefs that make
the simulation hypothesis possible.
But this is really astonishing, and in quasi-contradiction which
what you say above. We just don't know any phenomena which are not
Turing emulable.
But isn't that just a selection effect. If it weren't Turing
emulable, how would we know that?
We would not know it, but might infer it, like we infer the randomness
in the quantum, or in the iterative self-duplication.
As a theorician, but only as a theorician, I can show the
theoretical existence of non simulable phenomena, but that really
exists only in theory, or in mathematics. Worst, most non simulable
phenomena will be non distinguishable from randomness, and if we
are machine, we will never been able to recognize a non Turing
emulable phenomenon as such. It seems that the question is more
like "how can we believe something non Turing emulable could exist
in Nature".
But your argument assumes that arithmetic exists, which is also
"only in mathematics".
But arithmetic is simple and instantiated everywhere. I was just
saying that in nature we have to imagine ad hoc things to get non
computable phenomenon. e^iCt, with C = Chaitin numbers, we have a non
computable solution of the wave equation, but it is not a 'natural
phenomena', nor even recognizable from e^iRt, with R a random number.
After all, "Human brain is similar to the Nelder-Mead simplex
method. It often gets stuck in local optima."
That can happen. But I am not sure it can makes sense to doubt
about mechanism. You need to study hard mathematical theories to
even conceive non-comp. Non-comp seems possible in theory, and has
an important role in the epistemology of machines, but in nature
and physics, it simply does not exist.
Yet most people on the 'everything' list assume the universe is
infinite and uncomputable.
Not at all. The big whole is taken as simple, like all sets, or all
numbers. The UD is simple and Turing emulable, but the internal
perspective are not.
Isn't it implicit in Everett's multiverse (and even explicit in
Tegmark's)?
Everett universal wave is computable. Tegmark takes the whole of math,
but seems unaware that this cannot be a mathematical well defined
object, nor to take inyo account relative first person indeterminacies.
It might even be a reason to doubt comp, because comp might predict
the existence of more non computable phenomena that what we "see"
in nature (basically the personal outcome of self-superposition).
But as you say above, we wouldn't recognize them as non-computable -
except perhaps in the sense of random, as in quantum randomness.
If a white rabbit steal your computer, you might be less sure,
especially if the sky is full of pink elephants. But yes, you are
right, the non computable seems to lurk only in the background, as a
multiplier of histories, both in comp and Everett.
Also, the UD simulates not just the computable phenomena, but also
the non-computable, yet computable, with respect to oracles, and
this is even more complex to verify for a 'natural' phenomenon.
The winning physical histories/computations
What do you mean by "winning" and how do you know this?
By winning, I mean they sustains rich and stable intersubjective
agreement among populations of independent universal machines. And I
know this by UDA + thinking a bit. By being deep and linear, they
maximize the relative stability of the dreams, making them sharable.
Bruno
Brent
are those who are very long and deep, and are symmetrical and
linear at the bottom, apparently, but this must be extracted from
addition and multiplication, and it is partially done with the
gifts of distinguishing the truth (about a machine), and the many
modalities: the observable, the feelable, the communicable, the
provable, the believable, the knowable, etc (with reasonable modal
axiomatics and their arithmetical realization.
The ideally correct universal machine has a particularly rich and
intriguing theology, which is made refutable, because that theology
contains its physics. So we can compare with nature, and if comp is
false, we can measure our degree of non computationalism.
Best,
Bruno
--
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@googlegroups.com.
To unsubscribe from this group, send email to everything-list+unsubscr...@googlegroups.com
.
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
.
http://iridia.ulb.ac.be/~marchal/
--
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@googlegroups.com.
To unsubscribe from this group, send email to
everything-list+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en.