This direction is valid. I am writing a new book on physics for AIs, which requires dissecting our physics beliefs in such a way that an AI can understand them. And what I am finding is that our models only go so far and then we have to say "I don't know why this thing happens, I can only make a mathematical model that's as valid as I can guess".

When you start to learn physics, you learn what I call the Level 1 model. It is simplified and it leaves out complications. It's high school. Later at Level 2 you return to the topics and this time learn better math models for describing things and now they handle second-order complications but even those are still somewhat simplified. It's college. We advance to Level 3 and 4 where the math gets hard and tough and sometimes crazy, but it isn't always guaranteed to be complete - perhaps simplified solely to let us be able to solve the equations at all (QM for example) - nor even a full explanation, and it is all extrapolation to things we can't touch or perceive directly - for example "string theory". People at the doctoral level which is Level 5 and maybe 6 deal with the very tough complications in models. The problem, the universe is at Level 6, 7, 8 or worse. In the end we can't model it well at all. And worse, the people exploring this are often super-autistic, high IQ and a little crazy and some with monster egos. Talk about your basic Dr Strange, we're there. You can see this leads to issues.

The problem we encounter is that some of our theory is so esoteric we can't test it, as there is no way to touch the things it covers. You will never be able to 'see' a string, or directly see the Higgs field.

The Michelson-Morley experiment for the aether was based on flawed assumptions and models, it never really disproved the idea of the aether.

This direction is valid. I am writing a new book on physics for AIs, which requires dissecting our physics beliefs in such a way that an AI can understand them. And what I am finding is that our models only go so far and then we have to say "I don't know why this thing happens, I can only make a mathematical model that's as valid as I can guess".

When you start to learn physics, you learn what I call the Level 1 model. It is simplified and it leaves out complications. It's high school. Later at Level 2 you return to the topics and this time learn better math models for describing things and now they handle second-order complications but even those are still somewhat simplified. It's college. We advance to Level 3 and 4 where the math gets hard and tough and sometimes crazy, but it isn't always guaranteed to be complete nor even a full explanation, and it is all extrapolation to things we can't touch or perceive directly - for example "string theory". People at the doctoral level which is Level 5 and maybe 6 deal with the very tough complications in models. The problem, the universe is at Level 6, 7, 8 or worse. In the end we can't model it well at all. And worse, the people exploring this are often super-autistic, high IQ and a little crazy and some with monster egos. Talk about your basic Dr Strange, we're there. You can see this leads to issues.

The problem we encounter is that some of our theory is so esoteric we can't test it, as there is no way to touch the things it covers. You will never be able to 'see' a string, or directly see the Higgs field.

The Michelson-Morley experiment for the aether was based on flawed assumptions and models, it never really disproved the idea of the aether.

This direction is valid. I am writing a new book on physics for AIs, which requires dissecting our physics beliefs in such a way that an AI can understand them. And what I am finding is that our models only go so far and then we have to say "I don't know why this thing happens, I can only make a mathematical model that's as valid as I can guess".

When you start to learn physics, you learn what I call the Level 1 model. It is simplified and it leaves out complications. Later at Level 2 you return to the topics and this time learn better math models for describing things and now they handle second-order complications but even those are still somewhat simplified. We advance to Level 3 and 4 where the math gets hard and tough and sometimes crazy, but it isn't always guaranteed to be complete nor even a full explanation, and it is all extrapolation to things we can't touch or perceive directly - for example "string theory". People at the doctoral level which is Level 5 and maybe 6 deal with the very tough complications in models. The problem, the universe is at Level 6, 7, 8 or worse. In the end we can't model it well at all. And worse, the people exploring this are often super-autistic, high IQ and a little crazy and some with monster egos. You can see this leads to issues.

The problem we encounter is that some of our theory is so esoteric we can't test it, as there is no way to touch the things it covers. You will never be able to 'see' a string, or directly see the Higgs field.

The Michelson-Morley experiment for the aether was based on flawed assumptions and models, it never really disproved the idea of the aether.

This direction is valid. I am writing a new book on physics for AIs, which requires dissecting our physics beliefs in such a way that an AI can understand them. And what I am finding is that our models only go so far and then we have to say "I don't know why this thing happens, I can only make a mathematical model that's as valid as I can guess".

When you start to learn physics, you learn what I call the Level 1 model. It is simplified and it leaves out complications. Later at Level 2 you return to the topics and this time learn better math models for describing things and now they handle second-order complications but even those are still somewhat simplified. We advance to Level 3 and 4 where the math gets hard and tough and sometimes crazy, but it isn't always guaranteed to be complete nor even a full explanation, and it is all extrapolation to things we can't touch or perceive directly - for example "string theory". People at the doctoral level which is Level 5 and maybe 6 deal with the very tough complications in models. The problem, the universe is at Level 6, 7, 8 or worse. In the end we can't model it well at all.

The problem we encounter is that some of our theory is so esoteric we can't test it, as there is no way to touch the things it covers. You will never be able to 'see' a string, or directly see the Higgs field.

The Michelson-Morley experiment for the aether was based on flawed assumptions and models, it never really disproved the idea of the aether.

This direction is valid. I am writing a new book on physics for AIs, which requires dissecting our physics beliefs in such a way that an AI can understand them. And what I am finding is that our models only go so far and then we have to say "I don't know why this thing happens, I can only make a mathematical model that's as valid as I can guess".

When you start to learn physics, you learn what I call the Level 1 model. It is simplified and it leaves out complications. Later at Level 2 you learn better math models for describing things and now they handle second-order complications but even those are still somewhat simplified. We advance to Level 3 and 4 where the math get hard and tough and sometimes crazy, but it isn't always guaranteed to be complete nor even a full explanation, and it is all extrapolation to things we can't touch or perceived directly - for example "string theory".

The problem we encounter is that some of our theory is so esoteric we can't test it, as there is no way to touch the things it covers. You will never be able to 'see' a string, or directly see the Higgs field.

The Michelson-Morley experiment for the aether was based on flawed assumptions and models, it never really disproved the idea of the aether.