I just looked, for the first time, at the Glendinning electronic control system manual.
Overall, I'm encouraged at their use of CAN bus backbone for the communications link. CAN, if you don't know, is a automotive driven data system standard that has been in use for a couple decades now on millions of over the road vehicles of all types. It has inherent robustness in FEC (forward error correction), the electronic IC drivers at each end have good voltage transient withstand, and generally is a solid, proven physical layer.
\
I've done a little design work with it for my type of products.
So, I can't tell what Glendinning does for product testing, and the manual gives no indication. I can tell you what type of testing I would like to see in a system like this, and just the test plan would occupy pages of text. They have leveraged, fundamentally, like I said, a great data backbone, but product testing must go much further than that.
In a critical product space such as this, you need, as a manufacturer, to insure that the product has a low enough defect level to not cause your legal department to work overtime and drive you into bankruptcy with product liability judgements.
In a salt water environment, with "untrained" operators, with zero shipboard maintenance and testing protocols, with a wide variety of threats, the product design must be solid.
Here are some things that I would find interesting to see the results of, results that "fly by cable" will have zero issues with:
You are at sea, or even at dockside and a yacht pulls near you and starts transmitting on HF/SSB with a 1kW transmitter. Do you still have engine control? Plastic boats, unlike cars, have about zero inherent radio shielding characteristics.
You are in a bad sea, and the helm is flooded with seawater. Are the controls functional?
You have a person in the engine room with a 6W VHF HH radio, communicating with the fuel filler topside. Does the black box at the engine, on the other end of the data path, still function? Or does it want to put the transmisison into reverse?
There are a multitude of other scenarios. Car manufacturers, and the multi-billion $ airframe designers have done this testing. Has Glendinning?
PS; forgot to point out that Toyota, arguably the best in the business for product quality, had early toothing pains with electronic throttle controls. Do you remember 10 years ago or so the operator reports of uncommanded throttle operation?