On 12 Jul 2016, at 00:15, John Clark wrote:
On Mon, Jul 11, 2016, Bruno Marchal <[email protected]> wrote:
>> 1) Each copy saw only one city.
> Excellent! That is the correct 1-view description. Now, you
just need to interview each copy about the prediction made in
Helsinki and written in the diary to evaluate the better one.
How? Which turned out to be the better prediction, Moscow or
Washington? And was the prediction about John Clark or was it
about some mysterious figure named "you"?
The better prediction was "W v M and I have no clue which one".
In that case, both copies agrees with each other and all subsequent
similar experiences.
2) All the copies together saw 2 cities.
> Correct 3p description of the experiences of all copies. That
is the 3-1 view. We need it to get the correct "1)", but "all the
copies" is not a person,
Then asking the Helsinki Person "what city will you see?" or "how
many cities will you see?" is a nonsense question because this is a
world with people duplicating machines.
Yes, but it should be obvious to anyone understand the difference
between the 1p and the 3p views, that the 1p views are not duplicated
from the 1p view (the 1-1-view as opposed to the 3-1-views, and the
3-1-1 views) etc.
When the H-guy pushes on the button in Helsinki, he knows with
certainty (assuming computationalism and the protocol and the default
hypotheses) that such a guy will find itself in a box, in front of a
door, behind which only one city will be seen (in the 1p view).
> that is why you correctly add "together" (which is the 3-1
view, in which we are not interested).
I know, you're interested in "THE 1p view" but as you just
pointed out in a world with people duplicating machine "THE 1p
view" is meaningless, there is only "A 1p view".
Exactly, that is the root of the 1p indeterminacy. There will be at
all moments only one 1p view, from the points of you of all copies.
> we are asked about the 1-views.
You are asking about what one and only one city was seen
The question concerns the future, or the next state. Then, the
verification is asked to all copies, and those which are verified by
all copies, when discussing together for example, are the correct one.
by "the 1-views" and that is a incoherent question with no
coherent answer. Garbage in garbage out.
There is nothing incoherent, and indeed, those writing "W or M, and I
have no clue which one" all win. and all other prediction fails.
In the finite case, just one fail refutes the prediction. In the
infinite iteration of duplication, we can dismiss the negligible set
(in the analytical or computer-science theoretical sense).
>> 4) The statement "John Clark will see two cities"
turned out to be unambiguously true.
> In the 3-1 view, sure.
If they were logical it would be true from true from ANYBODIES
view, Helsinki man Moscow Man Washington man you name it;
Yes, and only "W v M" is true from anybodies views when of course they
keep in mind we are talking about the 1-views, and not the 3-1-views.
All the copies agree that they expected and eventually verified to see
only once city, and not knowing in advance which one.
John Clark will see two cities.
That is the 3-1-view.
As you are John Clark, you need to go out of your body to conceive it.
But to complete the thought experience, you need to re-integrate your
body after the duplication. As you have two bodies now, you have to do
a choice, or more seriously, you need to develop just enough empathy
toward BOTH copies, and listen to them: and both say that they see
only one city, and could not have guessed that one city in advance,
nor could they guess it again if we repeat the experience.
> But we asked about the 1-views.
There are 2 "1-views", and Bruno Marchal demands to know which ONE
and only ONE *you* will see, and that demand is pure gibberish.
You seem to be unable to understand that despite there are many 1-
views obtained, all the 1-views feel to be one individual in a
specific city.
By computationalism, you know that you will survive, and that you can
only feel to survive as a unique individual in only one city. You
*know* that in advance. The gibberish is only apparent to you because
you stop in the middle of the experience: you get first the correct
3-1 view, but you don't complete the thought experience, by, notably
looking at the personal memory of each copy, and remembering that the
question was about that "future personal memory". If you do that, you
can see easily that "W v M" win, and all others fail.
>> So which one was right?
> Trivially both when in Helsinki the prediction written in the
diary was "W v M",
But what exactly was the prediction about?
Given that the guy knows he will survive, and that he will feel to be
in one city, the question is "what city will be seen after pushing on
the button". The prediction is about one future first person experience.
If it was about how many cities John Clark will see there would be
universal agreement that answer turned out to be 2,
Again, that's the correct 3-1 view.
but if was about how many cities you will see there will never be
universal agreement on what the answer turned out to be because in a
world with people duplicating machines the personal pronoun used
will be ambiguous.
Not at all, as we have agree that all John Clark are John Clark, but
after the duplication, each John Clark will see only one city. So if
the question is "how many city will you see", the correct prediction
is 1. And indeed, after completion of the experience, all John Clark
agrees to be personally in front of only one city.
The problem is not the pronoun. The problem, for you, comes from the
fact that you get the correct 3-1 view, and then forget to just ask
all the copies about the verification of the prediction made in
Helsinki, and which has been duplicated together with the H-Guy.
A correct prediction is defined by a proposition which remains true in
all the cities where the H-guy survives (we work in the
computationalist theory).
The question is about the future personal memory, if you want.
In the finite duplication scenario, there are no difficulties at all.
No ambiguities, especially that we agree on the identity theory.
In the infinite scenario, and with composition of probabilities,
things are much more difficult, and that is why I use the self-
referential (Kleene theorems, Gödel-Löb theorems) to get "only" the
(quantum) logic for the P=0 and P≠0 scenarios, or the "yes-no
elementary case" (the propositional solution). It is hard enough.
Bruno
John K Clark
--
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 https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
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 https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
