Atoms make molecules.  Molecules make matter.  Though that's not entirely true it doesn't matter here.  Matter is anything that takes up space and has volume.  Stated differently, matter is something that has a practical component.  Language is imperfect, it follows the imperfect nature of human biology, it was after-all invented, not discovered.

"Science" is basically a human saying, "Maybe this is true" then saying "How about if we do {this} a bunch to test it".  It's not such a crazy notion, humans had already figured out a variety of safe dietary options and unsafe manners of behavior at this point - this was just the extension of human experience, informed by our nascent sense of empathy.  The guy who said "Lions are safe, we can cuddle with them," was destined to lose his vote. As did folks that irresponsibly ate known poisonous berries.

Civilization, and the knowledge behind it, is our bubble of protection against the violent state of nature.  This is the political philosophy behind all proper modern Democracies. As the American experience exists, the arc of philosophical evolution, like the arc of human justice, bends towards truth - as long as bad actors are prevented from disrupting it.

As knowledge became more complicated and we moved from "these fruits are good" to contemplating the nature of free-will, we needed an agreed upon system to adjudicate theorems.  This is called the "Scientific Method" and it leans towards more easily testable systems.  It's sensible.  The more complicated the conjecture, the more difficult in proving of it.

Occam's Razor is that the simplest explanation is the preferred one.  The quickest path between two points is typically thought to be a straight line.  My buddy Peterson used to love saying that.   

Maximum Entropy is the opposite of that.  A calculated Maximum Entropy for a hypothetical answer to a theorem is an array of all possible answers; wherein the theorem has sufficed Maximum Entropy, the aggregate opinion of entropy does not violate the base theorem.

Maximum Entropy is maximum complication.

The goal is to account for Maximum Entropy in your scientific tests.  It's a scientific flex; "I've accounted for all the craziest stuff that could possibly go wrong".  But entropy is exponential, you can only do so much.  Nature isn't just the planet, it's everything on it, in it and beyond it, including "things" humans don't even know about yet.

It's infinite against our limited human experience.  Thus our need for abstraction.  And while scientific ideology may drive the pursuit of maximum entropy, compromise will always be a component of it's expression.