Physics Intuition

I had an argument with a friend the other day. He’s quitting physics and will be teaching high school, which in my opinion is more of less a waste of talent.

§1 Complexity, or Why Most Physics Work is Writing Code

Here is our differences: He thinks most people in the physics business are doing something he does not like, which is writing code. For example, in his CP violation paper, what he mostly did is to get familiar with the code base. In his opinion, people should have physics intuition to come up with different theories and test it in experiments. Most of what we beginners do is the latter part, in other words, writing code. But my point is that physics has changed a lot and can not devolve back into the 20th century. The 21st century is about complexity:

• Theories that are complicated, intertwined and conflicted.
• Time: Experiments take decades now.
• Money: Newton did not need millions of dollars to buy lenses.
• With complexity comes with communication, which is inherently inefficient.
• Computers do so much more than line fitting, which is not achievable by human labor.

Compared with this list, writing code does not seem so bad.

I am sorry nature does not allow analytical solution for crystal lattice so that we use approximations and computer simulations, but there is just no other way.

Part of me blame the education system for a little bit: in undergrad, physics history in textbooks stops at mid-20th century.

BTW, in The Three Body Problem, if the Trisolarans write simulation program of their three body solar system, at least they can predict the movement of the solar system in the next decade or so. In the meantime, they can continue to observe the movement of the solar system and modify the CONSTANT of the program for better precision, such as the mass of each sun and their distances between each other. Also, in order to get better precision, they can replace Newtonian dynamics with General Relativity. Thus given their highly-advanced technology (near-light-speed travel and so on), there’s no need to leave their solar system and come to our solar system. Yes there’s no analytical solution for the three body problem, but simulation is really useful.

Even though I said decade or so in the last paragraph, I don’t know the specific time the program can predict, but I assume they evolved to intelligent beings thus must have a relatively long stable time. Maybe I can do a lot of simulations and take a mean value. And a more analytical method would be finding the saddle point in the system, as Poincaré did. Anyway, we only have one sun in the solar system so I’m not interested really.

§2 I Have Trust Issues (with My Physics Intuition)

Throughout our argument, one term he kept using is “comprehension”. I kept saying I would prefer “hypothesis”. I think this tells the difference between us. You see, I don’t trust my intuition at all. Because at the end of day what we do is guessing, systematically guessing through math yes, but still guessing. I am being so cynical mostly because of my experience, and Immanuel Kant. (BTW, I heard undergraduate philosophy education stops at Descartes, the 17th century.)

All physics intuitions are wrong, but some are more wrong. I read this article Physical Intuition vs. “Math” today, and the author wrote so many awesome examples.

I am so happy to announce: I have trust issues. And I don’t need intervention.

§3 Be Constructive

Just some notes for myself:

• Write code: You’re speaking three languages: the physics experiment language, programming language and physics theory language. Beware nobody else is trilingual, not even your professor, especially not your professor.
• Read papers. Not about the math though, more about ideas and directions.
• Show up to meetings and listen to people.
• Put physics models in your brain cache the whole time.