Yup, it sure is noisy. But let me indulge in my (not so realistic, but still potentially feasible) thought experiment :)
What if it turns out there is actually a predictable pattern to the noise on a ceramic surface. Perhaps there is a way that the molecules fit together that mean there is a few rules that govern the noise, such that we can then indeed remove it from any "recordings".
Or maybe there's some mathematical properties of noise that are yet to be understood which allow us to remove it correctly.
Or perhaps there is some arm of science which doesn't even exist yet (in the same way quantum computing didn't exist so long ago).