The LEDs in the keys have their cathode (ground) pins wired together in groups of eight. And we meticulously numbered every "power" side wire with little flags, so we'd know which wires to common together for the MAX7219 controller when it comes to driving the LEDs.
The whole thing took maybe a full day in total (Nick and I were out playing petanque/boules for most of Saturday daytime, and we had about eight hours at the soldering station on Sunday).
So now we've over a hundred solid-core(?!) wires stripped, soldered, cut-to-length and taped together (yes Steve, taped together with crappy old sellotape - no heatshrink here for us!) in a kind of wiring loom.
Some of the wires still need cutting to length, but we left them overly long, for testing on a breadboard. When it comes to it, we'll probably just solder the wires as they are and just stuff them all inside the case - after all, there'll be plenty of room inside now we've taken all that old circuitry out!
It's late now, so it'll probably be tomorrow (Monday) evening before we get any firmware running to test it all out. With a bit of luck, we'll have at least one of the two decks working fully before the next boogie-woogie band practice on Tuesday night, and we'll be able to try it out for real.