Or is it that as long as the substrate the nickel is plated on gets protected by the nickel from exposure to the electrolyte, it won't matter what the substrate is (as long as it conducts)?
Yes. The chemical reaction goes on where the electrode is in contact with the electrolyte, both for electroplating and the hydroxide formation process. So the underlying substrate can be made of anything, provided there is enough nickel on it to protect it from being exposed to the electrolyte. Since the long life of the Edison cell (as I understand it) is down to the nickel hydroxide being replenished from the underlying nickel during the recharging process, any surface plating will slowly be etched away as the battery goes through thousands of cycles.
As I understand it, the problem is how to get the nickel hydroxide, which is not a good conductor, into electrical contact with the terminal on the battery. This is what got Edison making flakes of nickel, and what is at the heart of the modern "pocket plate" design. If the nickel hydroxide layer is thin enough, this should not be a problem, but then we need a very large surface area if the battery is not to become prohibitively heavy.
Here are modern examples of ways to overcome this problem (in NiMH, but the nickel electrode is the same):-
http://www.freepatentsonline.com/4985318.html
http://www.electroenergyinc.com/products/technicalpapers/ScaleUp.pdf
http://gltrs.grc.nasa.gov/reports/2001/TM-2001-211068.pdf
These approaches are probably beyond the home constructor. Hence the wire wool, as it is probably the best compromise between surface area, conductivity, and kitchen sink technology.
If you're using nickel wire wool (still a possibility, if you're making lots of batteries and can order the stuff from the factory - it's called a "nickel felt electrode" - see P2 of the NASA report) then you don't have to worry about the surface plating breaking through, but can instead choose the thinnest grade of wire that will avoid the wires being broken by that same etching as cycles go by. But you still need to know how much Ni can be converted to Ni(OH)2 before you get too high an internal resistance.
If you're using plating, you'll need to experiment first to find out how much hydroxide gives the internal resistance you want; and then to find out how much more plating than this you need to get a reasonable cycle life.
To do that I'd use nickel wire, and try creating various depths of Ni(OH)2 and working out the internal resistance by discharge. Knowing the nickel wire radius gives the electrode area, and from that and the total capacity (in amp-seconds - amp-hours divided by 3600) divided by Faraday's Number gives you the number of moles per square metre. That should enable you to calculate how it will scale up.
Having calculated the amount of nickel you can convert to hydroxide per square metre, and the area of your electrode, you can work out how much nickel to plate to ensure that your plating doesn't break through when you form the hydroxide. But first you need the results of the wire experiments.
Edison did a lot of experiments before he had his cell working to his satisfaction. He also never got the environmentally friendly electric cars that were his dream. But there has been another ninety years of research into nickel hydroxide electrodes in alkaline solutions since then. Maybe Edison's dream can finally come true.
