Comments by "Lepi Doptera" (@lepidoptera9337) on "The Trouble with Many Worlds" video.

  1. 2
  2. 2
  3. 2
  4. 2
  5. 1
  6. 1
  7.  @jth4242  I didn't say that it's impossible to derive actual physical "elementary processes". What I said is that Feynman diagrams are not equivalent to these processes. That does not mean that there are no other representations of the behavior of quantum fields that can yield a much better interpretation. Personally I strongly suspect that there are and I gave you an example of recent theoretical insight into one possible such avenue. The problem with Feynman diagrams is simply that they weren't derived from physical insight but they are a purely mathematical necessity to get through the calculations in a naive way, at all. Why are Feynman diagrams like that? Because the path integral formulation of quantum mechanics, which is a quantization procedure, basically assumes that one can describe the world in terms of quanta, which are irreversible energy exchanges at the boundary of a quantum field. By pushing this irreversibility (that does not exist there) into the volume, we are opening up a whole slew of mathematical problems, which we then have to solve with more or less ad-hoc methods. Even our best solutions have remnants of trouble (like an extremely high calculated vacuum self-energy term) that do not exist in nature. In nature quantum fields behave like extremely smooth objects all the way across the universe (if they didn't, then we couldn't see all the way back to the beginning of time with telescopes). There are no dizzying virtual particles popping in and out of existence all over the place as poor visualizations of spacetime want to make you believe. Atoms in the ground state remain in the grounds state until actively excited etc... all of this tells you that something in our math is wrong. It's good enough to calculate a lot of things, but it's still not good enough to be satisfied with it. And we aren't. That's why very smart people are still thinking about the problem. QFT with Feynman diagrams is, by far, not the last word, neither mathematically nor in terms of physical interpretation.
    1
  8. 1
  9. 1
  10. 1
  11. 1
  12. 1
  13. 1
  14. 1
  15. @Imjust Observing Violation of any of the conservation laws. I don't know if you guys are aware of this (it does not seem you are), but an atom has to be brought into an excited state to release a photon. That means that a mixed state between exited and ground-state is simply some of the ensemble being in the excited state and some of it being in the ground state. The same atom, however, can never be in both states at the same time because that energy will, eventually, have to be released. Energy that is not in the system can not be released. Energy that is in the system can not stay in it forever because of the decay time constant. Goodness, gracious, kids. This ain't rocket science. One simply has to be able to distinguish between ensemble properties and individual system properties. What QM can't tell us for mixed states is which of the actual states one particular atom is in, that's all. This does have severe consequences for the phase space of systems that have many degrees of freedom, but even for those the total energy is the total energy, no matter how much it spreads out over the mix. You can make the same argument for any conserved quantity. Momentum, angular momentum, spin, charge, lepton number, whatever. These conservation laws are not just being obeyed in the average. They are being obeyed by any one of the samples in the ensemble. To think otherwise is unphysical. Maybe you really need to take a few lab classes in the atomic physics department to remind yourself how nature actually works. That is the problem with all this double slit nonsense and Schroedinger's cat etc. These toy examples do not force you to think about real physics. They all gloss over all the stuff that is of actual importance in real world experiments. Do you really think we would build a multi-billion dollar 17km circumference accelerator if protons would be equally likely in all possible momentum states????? Seriously? End rant.
    1
  16. 1
  17. 1
  18. 1
  19. 1
  20. 1
  21. 1
  22. 1
  23. 1
  24. 1
  25. 1
  26. 1
  27. 1
  28.  @2tehnik  All the measurement bullshit comes from people who didn't study relativistic quantum field theory or, if they did, lack the intuition to connect it to the fundamental questions of "reality" in quantum mechanics. Historically the study of quantum mechanical systems basically split at the end of the 1920s, shortly after the more or less final version of non-relativistic quantum mechanics was agreed upon. A number of physicists who were interested in low energy systems like solid matter and atomic and molecular spectra kept working with the non-relativistic theory, which can achieve reasonable accuracy for a number of problems but suffers from (mostly imaginary) interpretation issues because it is not a good theory of the world in general. It just happens to work if we squint and don't care too much about certain details. That's exactly how classical mechanics works in comparison to relativity. It's conceptually false but still good enough for many simple problems. The more serious faction of physicists like Dirac, however, realized immediately that non-relativistic QM was both insufficient and a kluge. They began to construct relativistic versions of the theory, initially very much in the style of the Schroedinger equation. These early attempts produced enormous mathematical problems and were very hard to work with. More importantly, they did not reproduce the plethora of ever higher energy physics data that came from ever more energetic accelerators easily. It took these physicists twenty years (all the way towards the late 1940s) to develop mathematical methods that could tackle non-trivial problems (like Feynman's path integrals and Feynman diagrams). It took another twenty years (bringing us closer to the 1970s) to tackle divergence problems and to develop what is called gauge theory that is the backbone of the standard model of physics. As a by-product of these mathematically very challenging problems the comparatively trivial issues of quantum mechanical measurement etc. are falling by the wayside completely. It is trivial to explain the source of uncertainty in a relativistic worldview and it is just as trivial to explain why the world seems to reduce to classical physics on its own without somebody staring at it all the time. So why are there still people who just can't take the relativistic quantum field theory "yes" for an answer? Beats me. Some people just don't want to get out of their Schroedinger comfort zone, I guess, no matter how poorly it matches to actual reality.
    1
  29. 1
  30. 1
  31. 1
  32. 1
  33. 1
  34. 1
  35. 1
  36. 1
  37. 1
  38. 1
  39. 1
  40. 1
  41. 1
  42. 1
  43. 1
  44. 1
  45. 1
  46. 1
  47. 1
  48. 1
  49. 1
  50. 1
  51. 1
  52. 1
  53. 1
  54. 1
  55. 1
  56. 1
  57. 1
  58. 1
  59. 1
  60. 1
  61. 1
  62. 1
  63. 1
  64. 1
  65. 1
  66. 1
  67. 1
  68. 1
  69. 1
  70. 1
  71. 1
  72. 1
  73. 1
  74. 1
  75.  @greghansen38  Scattering matrices don't have ontological problems. They are describing the scattering of waves from infinity with the outgoing solutions also being probed at infinity. This eliminates questions like "what is a measurement" completely (not that that's hard to answer in non-relativistic QM but one has to drag thermodynamics into the mix, which is not necessary in relativistic theory). It eliminates the classical-quantum system boundary. Infinity always provides more than enough degrees of freedom for decoherence without having to construct "measurement objects". Phase doesn't matter at infinity, it all automatically boils down to amplitudes, just to name a few advantages. I do agree with you all the way that we are not teaching classical scattering theory nearly enough. It is extremely important in practice (optics, electromagnetic systems engineering, radar, medical imaging etc.) but the average physics student gets to see almost nothing about it. I certainly didn't, except for a trivial 2-d Coulomb toy problem. We also didn't learn relativistic dynamics enough to have a solid handle on collisions and high-boost systems. That leaves accelerator physics mostly in the dark, even from a classical perspective. I was taught some nuclear physics, but basically all in the non-relativistic approximation. High energy physics? Forget about it. That, however, is the real "footprint" of the universe. OK, it would have been different at a university that had an active high energy physics program, but even those lectures (I took them later as part of my PhD) were kind of very basic and insufficient to understand QFT on more than a surface level. One can, of course, learn important lessons about QM from atomic, molecular and solid state physics, but the fact that we are teaching the formalism independently of its applications makes connecting the dots harder, IMHO. I hope that younger professors are slowly getting a better handle on how to give students a working knowledge of QM, because for sure mine didn't.
    1
  76. 1
  77. 1
  78. 1
  79. 1
  80. 1
  81. 1
  82. 1
  83. 1
  84. 1
  85. 1
  86. 1
  87. 1
  88. 1
  89. 1
  90. 1