3 ms·

Relevant Feynman: https://www.feynmanlectures.caltech.edu/II_25.html https://www.feynmanlectures.caltech.edu/II_25.html (At the bottom, cf. “unworldliness”) P

by turzmo 3y ago

Relevant Feynman: https://www.feynmanlectures.caltech.edu/II_25.html https://www.feynmanlectures.caltech.edu/II_25.html (At the bottom, cf. “unworldliness”) Physicist here, I don’t want to be too dismissive because reducing physics down to its basic principles is part of the game. However, I’m not sure this covers everything. We need a description or at least characterization of 4D Minkowski space. Continuity and even differentiability of functions is assumed. I also am not sure that canonical quantization is covered by this list. Quantizing the EM field, for example, is a big pain and not simply implied by the EM Lagrangian or even any quantization rules. I do not think you can go from this list to non-commuting Hermitian operators acting on a Hilbert space.

abnry 3y ago
The article is highly opinionated. This basically opens the floodgates for critical comments. But like you, I sense some value here, because distilling to fundamental principles is part of the game in physics. I am not knowledgeable enough to comment on how accurate it all is, but I am reminded of amazing things like Noether's theorem, where conservation principles come from symmetries. I would love to understand all of this better, and my assumption is that producing results from symmetries is what much of particle physics is focused on. I think I'm okay with how opinionated this article is because, while I am not qualified to judge, it gives me an incredible sense of the wonder of physics. And I think opinionated intuitions, albeit somewhat wrong or imprecise, are useful for guiding ones understanding.
the_seb 3y ago
Thanks for your input. Since you're knowledgeable about this, can I ask about the thermodynamics part? My understanding is that thermodynamics involve macroscopic laws that are based on microscopic laws. This is similar to how chemistry depends on physics. So why must thermodynamics be included in this list? Isn't all of macroscopic (thermodynamics) physics derived from the microscopic just like chemistry?
- krastanov 3y ago
  I will be a bit sloppy with my description here: There are two ways to "prove" thermodinamics. You can start from the axioms of probability and statistics *together* with an assumption that matter is made out of atoms or similar constituents, and you can derive, through statistical physics, most of thermodynamics. That is a perfectly reasonable and practical way to do it. However, it is extremely valuable to also understand that the vast majority of thermodynamics (including our understanding of entropy) can be derived from completely independent axioms that have no relationship to information theory or the knowledge of discreteness of matter. You need to define temperature as an transitive observable (zeroth law of thermodynamics), you need to define heat as a type of energy (first law of thermodynamics), and you need some version of the second law (there are a few equivalent ones, but the easiest is that "heat flows only from high temperature to low temperature"). Lastly, you need equations of state as a starting point (instead of assumptions about the microscoping constituents of matter). Tadaaa, you have all that is necessary to derive all our knowledge of thermodynamics without knowing statistics or atoms. Aesthetically this is incredibly pleasing and intellectually it is quite surprising that these two sets of axioms that look nothing alike produce the same self-consistent physics. To be productive in cutting edge research you usually need to be comfortable with both. Edit: corrected associative -> transitive as pointed out below.
  - lisper 3y ago
    > it is quite surprising that these two sets of axioms that look nothing alike produce the same self-consistent physics Why is that surprising? This sort of thing seems to be ubiquitous in our universe. F=ma is equivalent to the principle of least action. Turing machines are equivalent to the lambda calculus. Electrodynamics is equivalent to the existence of a finite reference velocity. Also: > You need to define temperature as an associative observable Did you mean transitive? My understanding is that the required axiom is that if system A is in thermal equilibrium with system B, and B is in equilibrium with C, then A will be in equilibrium with C. That's transitivity, not associativity.
    - krastanov 3y ago

by turzmo 3y ago

abnry 3y ago

The article is highly opinionated. This basically opens the floodgates for critical comments. But like you, I sense some value here, because distilling to fundamental principles is part of the game in physics. I am not knowledgeable enough to comment on how accurate it all is, but I am reminded of amazing things like Noether's theorem, where conservation principles come from symmetries. I would love to understand all of this better, and my assumption is that producing results from symmetries is what much of particle physics is focused on. I think I'm okay with how opinionated this article is because, while I am not qualified to judge, it gives me an incredible sense of the wonder of physics. And I think opinionated intuitions, albeit somewhat wrong or imprecise, are useful for guiding ones understanding.

the_seb 3y ago

Thanks for your input. Since you're knowledgeable about this, can I ask about the thermodynamics part? My understanding is that thermodynamics involve macroscopic laws that are based on microscopic laws. This is similar to how chemistry depends on physics. So why must thermodynamics be included in this list? Isn't all of macroscopic (thermodynamics) physics derived from the microscopic just like chemistry?

krastanov 3y ago
I will be a bit sloppy with my description here: There are two ways to "prove" thermodinamics. You can start from the axioms of probability and statistics *together* with an assumption that matter is made out of atoms or similar constituents, and you can derive, through statistical physics, most of thermodynamics. That is a perfectly reasonable and practical way to do it. However, it is extremely valuable to also understand that the vast majority of thermodynamics (including our understanding of entropy) can be derived from completely independent axioms that have no relationship to information theory or the knowledge of discreteness of matter. You need to define temperature as an transitive observable (zeroth law of thermodynamics), you need to define heat as a type of energy (first law of thermodynamics), and you need some version of the second law (there are a few equivalent ones, but the easiest is that "heat flows only from high temperature to low temperature"). Lastly, you need equations of state as a starting point (instead of assumptions about the microscoping constituents of matter). Tadaaa, you have all that is necessary to derive all our knowledge of thermodynamics without knowing statistics or atoms. Aesthetically this is incredibly pleasing and intellectually it is quite surprising that these two sets of axioms that look nothing alike produce the same self-consistent physics. To be productive in cutting edge research you usually need to be comfortable with both. Edit: corrected associative -> transitive as pointed out below.
- lisper 3y ago
  > it is quite surprising that these two sets of axioms that look nothing alike produce the same self-consistent physics Why is that surprising? This sort of thing seems to be ubiquitous in our universe. F=ma is equivalent to the principle of least action. Turing machines are equivalent to the lambda calculus. Electrodynamics is equivalent to the existence of a finite reference velocity. Also: > You need to define temperature as an associative observable Did you mean transitive? My understanding is that the required axiom is that if system A is in thermal equilibrium with system B, and B is in equilibrium with C, then A will be in equilibrium with C. That's transitivity, not associativity.
  - krastanov 3y ago