I think their point was that there is no empirical definition of information as it relates to the observer. The expurimint you cite worked upon a physical system that already had a state prior to the expurimint.
If everything is information, then nothing is.
A disordered system still has state. You just don't know what it is.
Fair point on the semantics. But I'm not talking about subjective 'knowledge.' I'm talking about the thermodynamic cost to maintain a state.
Landauer showed that information processing is physical (heat). I’m just extending that logic: if the universe has to process too much state data in one spot, the cost isn't just heat—it's lag (Time Dilation).
It doesn't matter if we observe the mess; the system still has to render it.