Essentially this one: terrencebakes.substack.com/p/jalapeno-c...

You're not sorry at all, Mike

Here you go!

But now I realize what I'm saying is oversimplifying. It's true for the weak force, but you wouldn't be able to see that the strong force confines from this point of view.

Yeah, same structure as massless KG theory.

Of course I realized I said something silly above... Maxwell theory is already free! I should have said (non-Abelian) Yang-Mills

The free part is also a product of gaussian integrals for Maxwell theory and GR!

Kind of the same idea as modeling an interacting quantum field theory by using a free theory!

There is a well-behaved limit where interactions are small enough that you can treat the real physical system as a hierarchy, where the 0th is non-interacting. en.wikipedia.org/wiki/BBGKY_h...

But this can become a matter of perspective rather than physics. For example if I consider the dynamics of a test mass moving in a fixed Kepler potential, then I have a preferred frame of the background potential. But I can promote this to the two-body problem where the big mass is also dynamical

You can make something precise: There are principal null directions determined by the Weyl tensor. I think the difficulty is figuring out if a direction is "preferred" if it depends on dynamical quantities rather than "background structure"

I agree it's tricky to define what "preferred frame" really means. It depends on whether you count frames arising from dynamical fields as "preferred" or not. This is not even specific to GR!

Maxwell theory beautifully has no preferred frame — Lorentz covariance built right in! It's possible to build a Lorentz-covariant theory that _does_ have a preferred frame, with e.g. a unit timelike vector field

IMO in a modern light, an aether means something that gives a preferred frame

What does 'subtantial' mean here? And what do we mean when we say 'aether'?

I missed the memo... And supposedly I'm one of the "core" devs 🙃

Also that https cert needs to be replaced!