Meaningful means it exists. By virtue of (variants of) the macroscopic decoherence interpretations of quantum mechanics and the best understanding I and three other long time rationalists have of cosmology, everything physically possible exists, either in a quantum mechanical branch or in another hubble volume.
To narrow it down a bit (but not conclusively) start out by eliminating all propositions that presuppose violation of conservation of energy, that should give you a head start.
Anything physically possible exists within our timeless universe-structure’s causal closures: When we talk “meaninful” or “not meaningful” we are really talking physics or not physics. Perpetual motion, for instance, isn’t physics. Neither is (as far as I know) faster than light travel or communication, reversing entropy, ontologically basic mental entities and a lot of other things. They do not exist in any world in our universe, thus not meaningful, not a thing you can experience.
This of course presupposes knowledge of physics… I’ll have to mull on that. Funny disagreeing with yourself while typing.
Perpetual motion, faster than light travel, etc. were falsified by scientific experimentation. This means that these hypotheses must have constrained anticipated experience. Maybe they are “meaningless” by some definition of the word (although not any with which I am familiar), but that is not the way Eliezer is using “meaningless”.
Would a powerful AI, from the “run_ai” is pressed on the command line till it knows practically everything ever give a significant probability to violation of conservation of energy?
Humans are really amazingly bad at thinking about physics, (Aristotle is a notable example, he practically formalized intuitive physics which are dead wrong,) but what if you aren’t?
I am nearly certain there exists some multiverse branch where humans study the avian migration patterns of the wild hog, but I too am nearly certain there is no multiverse branch within this mutiversal causal closure where even one electron spontaneously appears out of nothing and then goes on its merry way.
I agree this is a different viewpoint than a purely epistemological one, and that any epistemological agent can only approximate the function (defun exists-in-mutiverse-p...), but if you want be stringent, physics is the way.
Furthemore it patternmaches against my concept of how Tegmark invented his eponymous hypotheses: finding a basic premise and wondering if it is neccesary. Do we really need brains to talk about meaningful hypotheses, or do we just need a big universe.
Meaningful means it exists. By virtue of (variants of) the macroscopic decoherence interpretations of quantum mechanics and the best understanding I and three other long time rationalists have of cosmology, everything physically possible exists, either in a quantum mechanical branch or in another hubble volume.
To narrow it down a bit (but not conclusively) start out by eliminating all propositions that presuppose violation of conservation of energy, that should give you a head start.
Anything physically possible exists within our timeless universe-structure’s causal closures: When we talk “meaninful” or “not meaningful” we are really talking physics or not physics. Perpetual motion, for instance, isn’t physics. Neither is (as far as I know) faster than light travel or communication, reversing entropy, ontologically basic mental entities and a lot of other things. They do not exist in any world in our universe, thus not meaningful, not a thing you can experience.
This of course presupposes knowledge of physics… I’ll have to mull on that. Funny disagreeing with yourself while typing.
No, it doesn’t.
There is a hubble volume beyond ours where you agree with me.
There is a quantum branch where you agree with me.
I am not sure there is a distinction between the two.
Also, you are right, it feels inadequate.
Perpetual motion, faster than light travel, etc. were falsified by scientific experimentation. This means that these hypotheses must have constrained anticipated experience. Maybe they are “meaningless” by some definition of the word (although not any with which I am familiar), but that is not the way Eliezer is using “meaningless”.
Eliezer uses meaningless on belief networks. I know that. I have read most of the sequences.
See this
Says who? Even if your multiversal theory is right, that doens’t follow. Physics doens’t prove anything about the meaning of the word “meaning”.
Would a powerful AI, from the “run_ai” is pressed on the command line till it knows practically everything ever give a significant probability to violation of conservation of energy?
Humans are really amazingly bad at thinking about physics, (Aristotle is a notable example, he practically formalized intuitive physics which are dead wrong,) but what if you aren’t?
I am nearly certain there exists some multiverse branch where humans study the avian migration patterns of the wild hog, but I too am nearly certain there is no multiverse branch within this mutiversal causal closure where even one electron spontaneously appears out of nothing and then goes on its merry way.
I agree this is a different viewpoint than a purely epistemological one, and that any epistemological agent can only approximate the function
(defun exists-in-mutiverse-p...), but if you want be stringent, physics is the way.Furthemore it patternmaches against my concept of how Tegmark invented his eponymous hypotheses: finding a basic premise and wondering if it is neccesary. Do we really need brains to talk about meaningful hypotheses, or do we just need a big universe.
I don’t see how that addresses my comment. A sentence is meaningful or not because of the laws of language, not the laws of physics.