If you fail to recognize the ultimate openness of a relatively closed reality, you end up with data being relatively true to nature. That's ok in applied science where pragmatism rules, but in search for the fundamentals of reality, the ultimate ground on which the relative stands, then we have a huge problem. The second law of thermodynamics is such a fundamental law, saying energy is not to be generated in a closed system, and therefore, reality must come to an end when kinetic energy and potential energy outpowers each other so that universal expansion comes to a halt.
I reject that idea completely, suggesting the universe indeed grows as the one and only possible structure/field of perpetual motion. If true, that requires the total amount of energy to increase as to avoid a massive density loss in the system. Realizing energy grows would in fact answer all of our questions about reality. But as long as we're ruled by the 2:d law, questions will remain unanswered.
On-chip Maxwell’s demon as an information-powered refrigerator is a paper that supposedly argues in favour of the 2:d law. But if you read it right, there's good stuff pointing in the opposite direction. If you read it assuming:
- Universe is an open system, but never the less a solitaire
- Information, heat and energy are essentially of same qualities
- The papers "system" and "demon" are not separate, but two interacting aspects of the same system
- Low entropy equals high polarization of local energies (kinetic vs. potential)
You will note that the results all point towards energy/heat/information increasing in the overall system. Also note that "Constant external control parameters ng and Ng govern the System current and the coupling of the Demon to the System, respectively" p.1, is not what happens in non-experimental reality where Demon/System are functionally intertwined, and "constant external control" is impossible.