Take as long as you want. You might even want to take forever – remember that if you do, there’s probably a good reason your subconscious doesn’t buy the plan.

Newcomb briefly:

Without determinism, the predictor Omega is impossible. With free will, there is no fact of the matter which of the boxes the chooser will choose; therefore nothing to predict; therefore Omega will get it right 50% of the time modified by whatever nonsense ideology they happen to hold.

With determinism, the answer is obvious and simple.

If you have a paradox, you’ve done the logic wrong.

“…From this state of, in view of the fact that some of the known cosmological solutions seem to represent our world correctly, James Jeans has concluded that there is no reason to abandon the intuitive idea of an absolute time lapsing objectively. I do not think that the situation justifies this conclusion and am basing my opinions chiefly on the following facts and considerations.”

“There exist cosmological solutions of another kind than those known at present, to which the aforementioned procedure of defining an absolute time in not applicable, because the local times of the special observers used above cannot be fitted together in one world time. Nor can any other procedure which would accomplish this purpose exist for them; i.e., these worlds possess such properties of symmetry, that for each possible concept of simultaneity and succession there exists others which cannot be distinguished from it by any intrinsic properties, such as, e.g., a particular galactic system.”

“Consequently, the inference drawn above as to the non-objectivity of change doubtless applies at least in these worlds. Moreover it turns out that temporal conditions in these universes (at least in those referred to in the end of footnote 10) show other surprising features, strengthening further the idealistic viewpoint. Namely, by making a round trip on a rocket ship in a sufficiently wide curve, it is possible in these worlds to travel into any region of the past, present, and future, and back again, exactly as it is possible to in other worlds to travel to distant parts of space.”

In the interim, I’ll try to crunch some of the response backlog … (it might take some weekend time)

There’s also the assumptions about consciousness. Are simulations conscious? I have a theory of consciousness, it says that pure software simulations will not be, any more than a image of a rock hitting another isn’t really hard. Even if it also plays a nice cracking noise. My theory might be wrong, but at least I asked the question. If you don’t run science on the question, you’re statistically guaranteed to be wrong, and you can’t do that if you don’t ask.

The activation computer assumes determinism and plays with infinity. You can’t make an infinite computer, and adjusting it back to finity is likely to make profound, unsubtle changes to the logic.

“herefore, choosing to press “ACTIVATE” is equivalent to choosing to make it certain that Omega pressed “ACTIVATE” and almost certain that you are not Omega, but an inhabitant of one of the simulations launched by Omega’s actions.”

No. Either Omega has already pressed the button, or you are Omega. If they have, then it doesn’t ‘make’ Omega anything. Rather, in the unlikely event you could know anything about Omega, and you know Omega is like you, you can now safely conclude that Omega made a similar decision.

You determine that Omega is very likely to have already done so, no retroactive causation. If you instead push ‘cancel,’ you determine that Omega was always unlikely to have done so.

Though predicting real humans is a lot harder than this, so I suspect this account is missing most key details.

–

The paper is ignoring not only conciousness, but free will. Again, I have a theory of free will. It might be wrong but at least I can examine my assumptions.

This cancel/activate thing only makes sense as a causative event if you have free will.

If you don’t, you were always going to do one or the other, and the last one to find out is you. Rather than retroactively causing simulation, the simulation caused you to make a sub-simulation. Someone doing a meta-simulation of Omega could simulate – or change – the entire chain. They could fudge the results, too. So the argument only makes sense if free will is true.

But Omega doing the same relies on determinism being true, otherwise there’s an unavoidable random element. The chain will self-terminate before infinity, guaranteed.

Reasoning forward from contradictory premises leads strange places. In the real world, it’s clear that if Omega exists, we can’t know anything about them, and infinite Russian doll computers would require infinite energy and collapse into a black hole, followed by the entire universe. (Infinite mass => infinite grav force => infinite acceleration => cosmic apocalypse.)

–

TDT has this same flaw. (Page 78-79, reads as 82 of the pdf.) We can predict or retrodict that others like us will make similar decisions, but it doesn’t make anyone else do anything. Rather, others predict our decisions, and takes us into game theory when we try to predict their predictions.

We predict the calculators will display the same number because they have the same components arranged the same way, not due to one calculator somehow influencing the other. Om is, as far as we know, On, just somewhere else. Identical things behave identically.

Though if you can explain to me what’s timeless about TDT, I’m all ears.

Beane, Davoudi, and Savage at least come close, but ultimately they’re predicting that experiments won’t match theory, but blaming reality instead of the theory. This is sublimely anti-science.

A situation that would be bit-for-bit identical with having no creator. So creator=>true == creator=>false. Which means a property of no property.

If you can’t tell the difference between a simulation and the real world, you may as well be in a real world.

–

Okay, let’s assume a creator will contact us, but hasn’t yet.

Should you conclude something that has exactly no evidence for it? Or should you, given all available experience is identical to there being no creator, go with that?

Obviously, update when the creator contacts you. It may be contingent, but there’s no way to know it is contingent. Epistemology, ontology, meta, wrong question etc. You can’t know it is contingent, which means true == false. It is neither contingent nor not-contingent, you’re asking the wrong question.

–

Look at doing it wrong:

1.

Not a simulation, thinks you are in a simulation.
Suffers from not liking the idea of being in a simulation.
Must deal with world as real, despite beliefs.

Cannot learn better.

2.

Considers it unknowable. As above, this reduces to wrong question.

3.

In a simulation, identical to not being in a simulation, thinks they’re not.
Due to identicality, there are no bad consequences or bad predictions that result from this belief. Doesn’t mind the idea they’re in a real world. Must also treat world as real.

Can learn better if creator reveals themselves, and change their mind. Also, can use the revealed information to deal with the world as a simulation.

–

The evidence is that the world is real. The right mistake is to assume the world is real. I find it baffling that anyone fights the conclusion.

–

A theory with no consequences is not even a theory. The only prediction a simulation theory can make is when and how the creator will contact you.

If we assume, though, that the communications strategy of the simulator is the key to contact, and — relatedly — that we know nothing about the simulator prior to the initiation of communication, then quite clearly the prospect of coming to know whether we are, or not, ‘inside’ a simulation / creation is a contingent matter, subject to empirical revision (subsequent to a communication event). Anything we concluded about it in advance would be an unfounded assumption.

As an alternative to tapping our feet, or poring over sacred scriptures of revelation, it’s worth engaging with the Statistical Ontologists on these questions (Moravec, Bostrom, and Almond — in particular — all have sophisticated arguments of great relevance).