It's (sort of) discussing the work of Nima Arkani-Hamed and his Colleagues, and their attempt to reformulate quantum field theory so that it doesn't refer explicitely to space-time, in analogy to how the principle of least action of Lagrangian mechanics doesn't refer explicitly to the causality/determinism of Newton's formulation (modulo some subtitles). Here's a bit more about their technical work.

The motivation for doing this is as follows. When the principle of least action was discovered, it was philosophically very puzzling: why should this equivalent, non-causal formulation of newton exist? The reason of course turned out to be quantum mechanics! (it's a kind of stationary phase approximation to the path integral formulation of quantum theory, roughly speaking). Now in the early 21st century, there is a whole bunch of hints that space and possibly time are not truly fundamental quantities (in analogy to how newtonian causality was not fundamental). So how do you proceed?

Method 1: make random guesses.

Method 2: attempt to extend existing physics in non-random ways (e.g. string theory)

Method 3: take a lesson from history

Method 1 never works, method 2 is apparently too hard this time, so they are attempting method 3: if you need to get rid of space-time, don't MODIFY existing physics, but REFORMULATE it so that it doesn't depend explicitly on that concept, in analogy to how lagrangian mechanics reformulates newtownian mechanics to not depend explicitly on causality ( & quantum theory is then a simple deformation away). And if you can reformulate state of the art quantum field theories so they don't depend on space-time explicitly, perhaps a deeper theory will just be a simpler deformation away as well.

Wow, that's not as short, as I expected, but nailed it.

A TL;DR; of TL;DR; : they are talking about reformulating laws of physics, so that formulas would not be dependent on space and time.

Basically, the idea is that there are often multiple theories to try to explain reality, each of which is valid (though some have further-reaching explanatory power than others), and that, perhaps, there is not one theory which can fully explain reality (Grand Unified Theory). Perhaps reality defies such simple explanations, and instead will always produce multiple answers. The question, then, if there are always multiple answers is: what is the question?

Personally, I think the question is: "Do You love Me? Will You seek Me?"

TL;DR as far as I can tell is:

Each theory is pretty much based on a previous theory, with some aspect of the previous theory removed to make the theory a better assumption of reality. - A really cool example of this a scientist named Laplace presenting his work to Napoleon about the origins of the solar system. Previously, it was believed that God was involved somehow but his theory didn't include God. When Napoleon asked why God wasn't included, he said "I had no need for that hypothesis".

Einstein's equations are a pretty good way to explaining things but they break down in certain situations, just like Newton's equations did and Einstein's equations are really just Newton's equations extrapolated a bit more.

Scientists are now discovering ways of explaining the universe that doesn't require Einstein's aspect of space-time to describe motion but equally work quite as well as his theory of general and special relativity.