5 ms·

Physicist here. This is just self-congratulatory nonsense. Though I shouldn't be surprised, because I originally signed up for this site to comment against some

by knzhou 7y ago

Physicist here. This is just self-congratulatory nonsense. Though I shouldn't be surprised, because I originally signed up for this site to comment against some other self-congratulatory nonsense. Quantum mechanics is built on extraordinarily simple rules, in fact simpler ones than classical mechanics: all you do is sum up the amplitudes for various paths, and square it to get a probability. The problem is that the objects obeying these rules thus act neither like particles nor waves. So obsessing over whether quantum objects are "really" particles or waves is not useful. It certainly works sometimes. There are many limits where quantum objects behave just like particles, or just like waves (there must be, since we must be able to recover classical mechanics from quantum mechanics). And sometimes you can describe a quantum calculation in terms of classical wave or particle intuition, which is what's going on here. The double slit experiment is in fact so simple that I know of at least 5 ways to derive the result, using all sorts of different intuition. "Duane's law" is one of them, though I never knew it had a name, as it pops out rather straightforwardly from the mathematics. The reason that we don't try to extend this to cover all of quantum mechanics is because it doesn't work. Treating the quantum mechanics as if it were classical underneath leads to all kinds of paradoxes. Bell inequality violations tell us that the classical variables would have to either influence each other faster than light, or be pre-set since the beginning of the universe in a malicious conspiracy. The delayed choice quantum eraser tells us that if you want to treat the photon as a classical particle, you need to accept that it can retroactively change the past. Yet another failed attempt is quantum logic: if you insist on treating quantum superpositions as representing some unknown classical position of a particle, you need propositions about those positions to violate the rules of logic themselves. Hundreds of experiments probing quantum mechanics have been performed, all with the same results: committing to a classical ontology in terms of "waves" or "particles" costs you more than just accepting that quantum objects are different from either. Physicists have spent a lot of energy thinking about this. All of the author's claims were under active debate in the 1920s. He is just completely ignoring the following century of investigation, and claiming that it never happened!

lisper 7y ago
> committing to a classical ontology in terms of "waves" or "particles" is almost impossible Not that I disagree with this, but I'd like to hear what you have to say about de-Broglie-Bohm.
- knzhou 7y ago
  de Broglie-Bohm has the same problems as usual for such theories. I wrote an article once [0] about such things and I'll just quote from it. > To understand the working physicist’s position, it is useful to consider pilot wave theory. This alternative to the Copenhagen interpretation is often championed by realist philosophers, who would prefer that measurable properties, such as the position of a quantum particle, should always be well-defined. In this interpretation, the wavefunction is a classical field, like the electromagnetic field, called the pilot wave. Every particle always has a definite location, but it is “guided” by interaction with its pilot wave, allowing it to perform feats impossible for ordinary classical particles, such as quantum tunneling. Hence, the subtlety of quantum mechanics is dispelled by describing a quantum particle as both a classical particle and a classical field, in a literal interpretation of wave-particle duality. > For those who have struggled with the subtleties of quantum mechanics, this simple story sets off alarm bells. For one thing, where is the probability? If the electron really has a definite position, then why is it measured to be seemingly random, even after the electron settles into its lowest energy state? The answer is that the pilot wave is postulated to unpredictably shuffle about the location of the electron until its position reaches “quantum equilibrium” and matches with the predictions of the Copenhagen interpretation. This shell game is assumed to occur too quickly to detect. Worse, once it is over, the electron is predicted to hover in midair, perfectly still. Velocity measurements indicate otherwise, so pilot wave theory simply assumes they are all mistaken. Apparently, there is a real velocity, but it cannot be measured, and any attempt to do so yields something else. These ad hoc fixes allow pilot wave theory to avoid contradiction with the empirically verified uncertainty principle. > The pilot wave itself also has strange properties. We expect to be able to measure classical fields, but the pilot wave cannot be directly measured. Classical fields were introduced to avoid nonlocal “action at a distance”, but when the particle is measured, the pilot wave instantly collapses. The collapse is postulated to be faster than light but coincidentally completely undetectable, making pilot wave theory almost impossible to reconcile with relativity. Furthermore, “Bell test” experiments confirm that this problem cannot be removed in any refinement of the theory. > The reason most physicists are hesitant to accept pilot wave theory is that it appears to have the ether’s flaws. In exchange for the classical intuition of definite particle trajectories, the theory drastically increases the complexity of our world. It suggests many natural questions about the nature of the pilot wave and particle, then gives them unnatural answers which are hidden from observation. That is why, when physicists working on quantum foundations were polled at the 2011 conference Quantum Physics and the Nature of Reality, precisely zero vouched for the pilot wave. The interpretation is like an art teacher who sings the praises of creative freedom, yet berates any who draw outside the lines. I don't personally have anything against Bohmian mechanics except for one thing: its supporters consistently ignore all of these well-known issues and present it as if it's a completely perfect quasi-classical model for the world, only kept down by conspiracy. This leads to all kinds of crazy narratives that I think are actively harmful to the field. At the very least, people who go out and talk about the many worlds interpretation face the weirdness of quantum mechanics head on, and honestly. 0: https://knzhou.github.io/#writing https://knzhou.github.io/#writing
ryanthedev 7y ago
Here we go. Throwing around internet titles. You do realize he has a PHD in math. I will trust his math work over your "theories" Let's see some papers you have published?
- knzhou 7y ago
  You can check my website, which I linked above, for several papers and a dissertation. I'm not talking about any of "my" "theories" at all, I'm just describing a bit of what has been done in the past century by thousands of smart and dedicated people.
- dang 7y ago
  Please don't take HN threads further into flamewar. Whatever the substantive points on the topic are, posting like this just makes the thread considerably worse. https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html
didgeoridoo 7y ago
> the classical variables would have to either influence each other faster than light, or be pre-set since the beginning of the universe in a malicious conspiracy Why is superdeterminism almost universally rejected by working physicists? From the outside, it feels like the most parsimonious explanation: there is just a single way things are and will be, set in motion from the very beginning. It doesn’t feel like a conspiracy of any kind to me. I’m definitely missing something, but what?
- knzhou 7y ago
  Because if you take superdeterminism seriously, you can't do science. Sure, it is logically possible that the world runs on classical mechanics, but that random factors in the light emitted from opposite ends of the observable universe billions of years ago were delicately set at that point so that whenever you run an experiment with it, naively assuming that no such correlations exist, quantum mechanics appears to be true instead. Unfortunately, if you are consistent with applying this principle you can't conclude anything at all. If you see the Sun and planets move every day, would you eventually accept orbital mechanics, or would you prefer the explanation that "orbital mechanics is false, everything actually moves however it wants according to random unknown factors, but those factors were arranged just so that whenever we make observations, standard orbital mechanics appears to be true"? Suppose we're betting on horce races, but I win 100 times in a row while you always lose. Would you accept my explanation that "this was merely superdetermined to be the case"? Or would you start to suspect that I'm stealing your money? Superdeterminism is philosophically equivalent to Cartesian skepticism. It's just a minor twist on being a brain in a vat. Scientists reject this not because we can disprove it, but because you can't do anything if it's true. I mean, you might as well not bother getting out of bed in the morning.
  - ncmncm 7y ago
    > you might as well not bother getting out of bed in the morning. Sometimes I feel that way. But I get up, take my meds, and press on.
- jeremysalwen 7y ago

by knzhou 7y ago

lisper 7y ago

> committing to a classical ontology in terms of "waves" or "particles" is almost impossible Not that I disagree with this, but I'd like to hear what you have to say about de-Broglie-Bohm.

knzhou 7y ago
de Broglie-Bohm has the same problems as usual for such theories. I wrote an article once [0] about such things and I'll just quote from it. > To understand the working physicist’s position, it is useful to consider pilot wave theory. This alternative to the Copenhagen interpretation is often championed by realist philosophers, who would prefer that measurable properties, such as the position of a quantum particle, should always be well-defined. In this interpretation, the wavefunction is a classical field, like the electromagnetic field, called the pilot wave. Every particle always has a definite location, but it is “guided” by interaction with its pilot wave, allowing it to perform feats impossible for ordinary classical particles, such as quantum tunneling. Hence, the subtlety of quantum mechanics is dispelled by describing a quantum particle as both a classical particle and a classical field, in a literal interpretation of wave-particle duality. > For those who have struggled with the subtleties of quantum mechanics, this simple story sets off alarm bells. For one thing, where is the probability? If the electron really has a definite position, then why is it measured to be seemingly random, even after the electron settles into its lowest energy state? The answer is that the pilot wave is postulated to unpredictably shuffle about the location of the electron until its position reaches “quantum equilibrium” and matches with the predictions of the Copenhagen interpretation. This shell game is assumed to occur too quickly to detect. Worse, once it is over, the electron is predicted to hover in midair, perfectly still. Velocity measurements indicate otherwise, so pilot wave theory simply assumes they are all mistaken. Apparently, there is a real velocity, but it cannot be measured, and any attempt to do so yields something else. These ad hoc fixes allow pilot wave theory to avoid contradiction with the empirically verified uncertainty principle. > The pilot wave itself also has strange properties. We expect to be able to measure classical fields, but the pilot wave cannot be directly measured. Classical fields were introduced to avoid nonlocal “action at a distance”, but when the particle is measured, the pilot wave instantly collapses. The collapse is postulated to be faster than light but coincidentally completely undetectable, making pilot wave theory almost impossible to reconcile with relativity. Furthermore, “Bell test” experiments confirm that this problem cannot be removed in any refinement of the theory. > The reason most physicists are hesitant to accept pilot wave theory is that it appears to have the ether’s flaws. In exchange for the classical intuition of definite particle trajectories, the theory drastically increases the complexity of our world. It suggests many natural questions about the nature of the pilot wave and particle, then gives them unnatural answers which are hidden from observation. That is why, when physicists working on quantum foundations were polled at the 2011 conference Quantum Physics and the Nature of Reality, precisely zero vouched for the pilot wave. The interpretation is like an art teacher who sings the praises of creative freedom, yet berates any who draw outside the lines. I don't personally have anything against Bohmian mechanics except for one thing: its supporters consistently ignore all of these well-known issues and present it as if it's a completely perfect quasi-classical model for the world, only kept down by conspiracy. This leads to all kinds of crazy narratives that I think are actively harmful to the field. At the very least, people who go out and talk about the many worlds interpretation face the weirdness of quantum mechanics head on, and honestly. 0: https://knzhou.github.io/#writing https://knzhou.github.io/#writing

ryanthedev 7y ago

Here we go. Throwing around internet titles. You do realize he has a PHD in math. I will trust his math work over your "theories" Let's see some papers you have published?

knzhou 7y ago
You can check my website, which I linked above, for several papers and a dissertation. I'm not talking about any of "my" "theories" at all, I'm just describing a bit of what has been done in the past century by thousands of smart and dedicated people.
dang 7y ago
Please don't take HN threads further into flamewar. Whatever the substantive points on the topic are, posting like this just makes the thread considerably worse. https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html

didgeoridoo 7y ago

> the classical variables would have to either influence each other faster than light, or be pre-set since the beginning of the universe in a malicious conspiracy Why is superdeterminism almost universally rejected by working physicists? From the outside, it feels like the most parsimonious explanation: there is just a single way things are and will be, set in motion from the very beginning. It doesn’t feel like a conspiracy of any kind to me. I’m definitely missing something, but what?

knzhou 7y ago
Because if you take superdeterminism seriously, you can't do science. Sure, it is logically possible that the world runs on classical mechanics, but that random factors in the light emitted from opposite ends of the observable universe billions of years ago were delicately set at that point so that whenever you run an experiment with it, naively assuming that no such correlations exist, quantum mechanics appears to be true instead. Unfortunately, if you are consistent with applying this principle you can't conclude anything at all. If you see the Sun and planets move every day, would you eventually accept orbital mechanics, or would you prefer the explanation that "orbital mechanics is false, everything actually moves however it wants according to random unknown factors, but those factors were arranged just so that whenever we make observations, standard orbital mechanics appears to be true"? Suppose we're betting on horce races, but I win 100 times in a row while you always lose. Would you accept my explanation that "this was merely superdetermined to be the case"? Or would you start to suspect that I'm stealing your money? Superdeterminism is philosophically equivalent to Cartesian skepticism. It's just a minor twist on being a brain in a vat. Scientists reject this not because we can disprove it, but because you can't do anything if it's true. I mean, you might as well not bother getting out of bed in the morning.
- ncmncm 7y ago
  > you might as well not bother getting out of bed in the morning. Sometimes I feel that way. But I get up, take my meds, and press on.
jeremysalwen 7y ago