On 22 Nov 2012, at 15:55, John Clark wrote:
Bruno Marchal <[email protected]> wrote:
> ?
!
> by comp you know that you will survive and feel to be unique in
once city,
By comp who knows that what will survive and feel to be unique in
one city?
Any human (and Löbian machine) can understand that if comp is true, it
means that they will survive and feel unique in one city. That is
confirmed by the interview of the resulting copies.
> you cannot predict which one in particular.
What cannot predict which what in particular?
Nobody can predict in which city anyone can feel to be in a WM-
duplication, even God.
If God predicts W, the M-guy will understand that God failed.
If God predicts M, the W-guy will understand that God failed.
If God predicts M & W, both guys will understand that God failed.
Only choice: W or M. In that case, both copies can confirm that God
was correct.
>> I can conceive of 2 things being identical from the 1-view but
not from the 3-view, the biological John Clark and the uploaded John
Clark for example; but I can not conceive of 2 things being
identical from the 3 view but not from the 1-view,
> That simply never happens.
It violates no laws of physics and happens all the time in the world
that you created, the world where duplicating chambers are
ubiquitous. If 2 bodies and brains are identical from the 3-view
their subjectivity is identical from their own view. If you dispute
this give me a counterexample.
I have never dispute this. The question concerns the result of the
differentiation brought by the self-localization done after the
duplication.
> you keep saying that after the duplication you will be in both
places.
John Clark keeps saying that after the duplication John Clark will
be in both places.
Not from its personal subjective view (1p). The question bear on this.
> That is correct in the 3p view, but non sensical in the 1p views.
Where subjectivity is concerned if things are identical from a third
person perspective they are certainly identical from any first party
one. If you think I'm wrong give me a counterexample.
They are not identical. One body is reconstituted in W, and the other
in M.
> With QM I am multiplied in more than 10^100+ at each instant,
John Clark is and Bruno Marchal is, but as for "I" that depends on
who "I" is.
So you agree now that the situation is equivalent, with respect to
probability, for QM and comp?
> you are only one person, in all situation
Who is only one person in all situations.
All person when they refer to their subjective 1p.
> there is only one first person view.
There are over 7 billion first person views on this planet, and
growing.
Yes, but they are all unique from their first person point of view.
> You know that you will survive in one city.
Who knows that what will survive in one city?
See above.
>> WHICH WHAT? WHICH YOU??
> Which city, and which first person experience
Which first person experience?? Until one duplicate sees Moscow and
the other duplicate does not the two brains produce first person
experiences that neither a third party nor the first party itself
can distinguish between; and if there is only one thing it's
nonsensical to ask "which one?" about anything.
Again, what you say is obvious, and non relevant. the question is
asked to the H-man, about which city he will see after pushing the
button, *and* opening the door (localizing himself).
> the question is about which change you will live.
Which change who will live?
Who? Anyone doing the experience fully described in step 3.
> you only confuse the 1-you with the 3-you.
John Clark believes it is Bruno Marchal that is confused because
Bruno Marchal seems to believe that it is easier to produce things
that are identical from the 3 view than from the 1 view, when in
fact the opposite is true.
> you self-localize.
"You" don't do anything, the location does something to "you", the
city causes a change in "you" that is not made in your no longer
exact twin so now a third party could distinguish between the two,
and now it's possible to do what could not be done before "you" can
tell that its twin is not "you".
And so a selection has occured, let us say in Moscow, and the H-guy
knew this in advance, so the question asked in Helsinki made sense,
and he understand that he could have guess in advance that he would
have been the M-guy.
Nor can anybody predict a WM-sequences in the iterated case, as most
of them are random-incompressible.
> You might consider a protocol where the person ignore if the
protocol will be able or if he will be send (under anesthesia) after
a coin choice between W and M. You talk like if you could
distinguish the two protocols,
I don't even understand what you're saying, read the above and see
if you do.
OK, I was unclear.
Consider the following two protocols:
1) In Helsinki you are put under anesthesia, then sent to W or to M,
according to a random coin choice. You wake up in a room, open the
door and see that you are in a city.
2) same protocol as in step 3, except that the annihilation is done
when you are under anesthesia.
And now consider the following third protocol. You are in Helsinki,
and you are told that we will repeat ten times, either the protocol 1,
or the protocol 2, but without telling you which one.
The question is, can anybody coming back from such an experience
decide if the protocol 1 has been used, or if the protocol 2 has been
used. We suppose that in case the protocol 2 has been chosen, the
copies have no means to see each other.
> The notion of Löbian machine is completely standard,
About all Google knows about "Löbian machines" is what's said about
them on this very list, mostly by you.
I use that name in place of "sufficiently strong theory" which is
often used in the literature.
The very fact that you are doing remark like that show that you agenda
is just being negative about me, and that you are not interested
neither in the mind-body problem, and still less in the comp mind-body
problem and its reduction to a body problem in arithmetic.
I call them Löbian because it can be proven that their probability
predicate obey to Löb's formula ([]([]p->p)->[]p). It means that theor
beliefs are close for the Löb's rule: if ever they prove []p->p, for
some p, they will believe p.
That formula is counter-intuitive, but by replacing p with f, or with
"0= 1" you can easily see (if you know propositional calculus) that it
entails Gödel's second incompleteness theorem. More importantly,
Solovay's theorem shows that the Löb formula is the main axiom of the
modal logic characterizing the provability predicate of all theories
or machine whose theorems or beliefs extends effectively the theorem
of Peano Arithmetic (say). I model "believability" by "provability",
which makes sense in the case of ideally correct machines.
>> Ed, after the duplication will Ed see Washington?
Yes.
Ed, after the duplication will Ed see Moscow?
Yes.
Ed, after the duplication will Ed see one city and one city only?
Yes.
Ed, is there a 100% chance that all of that will happen?
Yes.
Ed, is that a contradiction?
No it is not because ED HAS BEEN DUPLICATED.
> You confuse the 3-Ed, and the 1-Ed,
If I've got the 3-Ed right the 1-Ed is too, and the 3-Ed looks right
to me.
If you got the 3-Ed right, you might have got the 3-view on the 1-Ed
right, but not their personal 1-view on those 1-view, as they are
unique.
> "W" represents "the first person experience
Who's first person experience?
Of the guy opening the door and seeing W. That is one of the Helsinki
guy, after he pushed on the button and did the self-localization.
>> Simultaneously ED has a first person experience of seeing W and
only W, AND Ed has a first person experience of seeing M and only M.
> Correct 3p view on the 1-Ed.
Please give John Clark an example of a correct 3p view of the 1-Ed
but the incorrect 1-Ed from Ed's own point of view.
"John Clark feels to be only in M". This is not correct for the 1-Ed
feeling to be only in W.
>>> evaluate P(W) and P(M)?
>> 100% and 100%.
> You again forget that "W" is for the subjective experience of
being in W.
I've forgotten nothing and the probability the ED will have the
subjective experience of being in Washington is indeed 100%.
This is refuted by the 1-Ed in M.
he can decide to ascribe consciousness, and even Ed-identity to the
doppelganger, but that is not the same as his current personal
feeling, and the question was about that person self-localization
feeling.
I see only hand waving to deny the obvious.
> both copies will have to say I was wrong
Ed will say that he (Ed) is experiencing Washington just as Ed
predicted and Ed will say that the fellow in Moscow has as much a
right to call himself "Ed" as I (Ed) do and he (Ed) is experiencing
Moscow exactly as Ed predicted.
But the question was not about the felling of all Ed, but on all
personal account to it. You confuse, like in the first post:
-"I feel to be in M and I know intellectually that my doppelganger
feels to be in W" with
-"I feel to be in W and I know intellectually that my doppelganger
feels to be in M"
The guy in Helsinki knows in advance that such feeling are
incompatible, exclusive.
> We have already agreed that both the W-man and the M-man can be
considered as both genuine survivors of the H-man.
And yet Bruno Marchal insists the H-man will not see W and M.
I say that from the first person perspective the H-man will see W, or
will see M. He will not see both W and M, from the first person
perspective.
It looks like you just don't hear the simple question asked.
Bruno
Two different things are happening to two different things and
because that can't be described as one thing happening to one thing
Bruno Marchal is claiming a new and profound type of indeterminacy.
John K Clark
--
You received this message because you are subscribed to the Google
Groups "Everything List" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to [email protected]
.
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 [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en.
