When one uses an existing word, it carries a lot of epistemic and ontological baggage, thereby causing trouble to those who are new to the domain. That’s why scientists go for a linguistic reform: invent a new word during a conceptual or scientific revolution. That’s how the word “Oxygen” replaced the old “Phlogiston”.
It even predicts a direct equivalence between the unit of bit and Joules per Kelvin with the entropy produced in the erasure of a bit being equal to 9.6e-24 J/K
Unless I'm making some mistake, you can even use computer science principles combined with Landauer's limit to solve Maxwell's Demon.
My least favorite example of this is statistical "significance" which gives the impression of importance. I can't think of a word that has caused more trouble in science than that.
