+1. As someone who has worked in the field of information theory, I strongly dislike the abuse of its usage here.
Funny that the term "information theoretic" is used with not even one "theorem" in the linked document.
Has anyone offered a better patch? I'm genuinely curious and don't know.
A patch [0] that added Fortuna [1] support to Linux was offered over 10 years ago.
Fortuna specifically addresses the problems of entropy estimation that the grandparent comment brings up:
"making any kind of estimate of the amount of entropy is extremely difficult, if not impossible. It depends heavily on how much the attacker knows or can know, but that information is not available to the developers during the design phase."
"Fortuna solves the problem of having to define entropy estimators by getting rid of them."
1: https://www.schneier.com/cryptography/paperfiles/fortuna.pdf
What's "a better patch"? We need to have a shared understanding of the problem before we can evaluate solutions.
To the extent that any applications exist where a information theoretic RNG was required, this scheme does not provide one. Instead, it continues the to offer an interface that causes confusion and insecurity without an apparent benefit.