It's interesting, but I'm at a loss as to what's even remotely new about it. Unless the computer that you typed this on is very, very different from mine that I'm typing this on, there isn't any living components in it, yet it is "computing" and "deciding" quite a bit. There's a ton of natural processes that incorporate some feedback element and can be said without much stretching to be doing a computation. We're beyond science on that matter, we're decades if not centuries into engineering with these facts. (It's not hard to say a steam engine is "computing" parameters to keep itself running, for instance, and h

Maybe it's just my bias as a computer scientist, but I would never dream that they would find anything but what they found. Having read the paper, they're basically proposing alternate models of computations, but computer scientists have hardly been blinded by transistors; proposing alternate models is a hobby, and there's at least one large one getting a lot of study to the point that I expect everyone will just understand the acronym without my expansion, QC. We've computed with water (both macroscopic and microfluidics), mechanical machinery (i.e., cogs not transistors), chemistry, DNA, analog circuitry, and light, and I'm sure that's not a complete list. We've hypothesized computing with mechanical nanotechnology, von Neuman replicators (up to and including converting entire astronomical bodies), black holes, closed timelike curves, and the fundamental structure of spacetime itself. If this was proposed as a Master's thesis in computer science, the advisor would advise the student to do something less pedestrian.

That doesn't make this paper "bad" in some absolute sense, but I'm surprised it's publishable, since for better or worse that incorporates a certain amount of novelty in its criteria.