5 ms·

A lot of complexity is hidden behind the term “information”. You should be careful not to just use your existing intuition/definition for this word, it’s extrem

A lot of complexity is hidden behind the term “information”. You should be careful not to just use your existing intuition/definition for this word, it’s extremely specific Quantum Mechanics jargon here. This is talking about quantum states and how they describe the world. Each state corresponds to physical (quantum) reality, conforming to the laws of physics. So it’s not like you can just twiddle bits to make new representations. I think this is an area where appealing to the lay reader’s intuition is counterproductive. If you haven’t solved the Schrodinger equation before then you definitely shouldn’t be trying to intuit things about quantum systems; they are just weird and kind of irreducibly complex from the mathematical representation. Let me attempt to go against my advice above and give you some intuition for why encryption doesn’t parse here. It would be like you have a program with some static types, some classes, and then say “what if we just encrypt the memory location for this object on the heap and run the program”. The program is the thing that is running (laws of physics), the variables on the stack/heap are the state for the current execution, and it has no concept of decryption, so it would just produce garbage and crash. In the same way, the quantum physics description of a system has superposed states that are all valid configurations of the physical system, and no notion of “encryption”. So there is nowhere in the model of physical reality (and therefore unless we are missing some new Physics, nowhere in the reality that is modeled) for this information to “hide”. Or taking a different tack, “thermal entropy” means it’s just a bunch of gas buzzing around randomly at the same temperature - there is no physical place for structure to be “encrypted”. Where is the “key” in your model of the world? It’s just a cloud of gas. What physical process performed the encryption? That would require a complex structure, yet we are talking about a cloud of particles emitted when one half of a particle-antiparticle pair is captured by the black hole’s event horizon. There is no place in a workable physical model of the world for an entity that performs encryption on the quantum states (whatever that might mean). All this just points to why you can’t encrypt states in this way, not why the black hole information paradox is a problem. For that you really do need the maths; eg see https://www.cs.umd.edu/class/fall2018/cmsc657/projects/group_2.pdf https://www.cs.umd.edu/class/fall2018/cmsc657/projects/group... for the Physics here; while that requires graduate-level understanding of QM, hopefully the intro will be useful.

lupire 5y ago
Obviously no idea if this is true, but, an object falling into a black hole maybe could emit some radiation containing the key before the majority of the mass "encrypted" into the black hole.
prewett 5y ago
> If you haven’t solved the Schrodinger equation before Heh heh, define "solved"... If I remember my QM class correctly, we "solved" the Schrodinger equation for the hydrogen atom; the book achieved this by observing "... which gives us <equation>, and, hey! look at this! it turns out that FamousLastName polynomials--which you've never heard of--turn out to solve this equation, here's their definition, and... problem solved!" (If I remember correctly they were Legendre Polynomials, but maybe that was some other equation. And to be fair to the book, it was a pretty good book.) After having "solved" the equation for one isolated, most-basic atom, they went on to say, "basically we have no idea if there is an analytic solution for anything more complex". If manipulating bra and ket vectors symbolically around an equals sign counts, we did a lot of that, although it did not develop in me the least intuition about QM. But then, I never really understood what those bra and kets were doing, and my grades steadily dropped (fortunately for my GPA there were only three courses). So it's possible I might have developed some intuition had I understood what was going on.
Ar-Curunir 5y ago
> the encryption ... would require a complex structure Not a quantum information theorist, but am a scientist. We actually do have (conjectured) low-complexity one way functions, so this by itself is not necessarily true. I do agree that it's a fairly unlikely to be the case that natural processes execute this algorithm, though.
- theptip 5y ago
  I think the GP is thinking of two-way encryption under a symmetric key here, else it’s hard to see how the information isn’t still “lost”.

lupire 5y ago

Obviously no idea if this is true, but, an object falling into a black hole maybe could emit some radiation containing the key before the majority of the mass "encrypted" into the black hole.

prewett 5y ago

> If you haven’t solved the Schrodinger equation before Heh heh, define "solved"... If I remember my QM class correctly, we "solved" the Schrodinger equation for the hydrogen atom; the book achieved this by observing "... which gives us <equation>, and, hey! look at this! it turns out that FamousLastName polynomials--which you've never heard of--turn out to solve this equation, here's their definition, and... problem solved!" (If I remember correctly they were Legendre Polynomials, but maybe that was some other equation. And to be fair to the book, it was a pretty good book.) After having "solved" the equation for one isolated, most-basic atom, they went on to say, "basically we have no idea if there is an analytic solution for anything more complex". If manipulating bra and ket vectors symbolically around an equals sign counts, we did a lot of that, although it did not develop in me the least intuition about QM. But then, I never really understood what those bra and kets were doing, and my grades steadily dropped (fortunately for my GPA there were only three courses). So it's possible I might have developed some intuition had I understood what was going on.

Ar-Curunir 5y ago

> the encryption ... would require a complex structure Not a quantum information theorist, but am a scientist. We actually do have (conjectured) low-complexity one way functions, so this by itself is not necessarily true. I do agree that it's a fairly unlikely to be the case that natural processes execute this algorithm, though.

theptip 5y ago
I think the GP is thinking of two-way encryption under a symmetric key here, else it’s hard to see how the information isn’t still “lost”.