Preface: This post ended up being jam-packed with technical details and jargon – a bit of a “devlog” feeling, if you will. If anyone wonders anything, ask away! No such thing as a stupid question! 
The package @Akuji (so colossally kindly) sent has (so eminently excitingly) arrived!
I’ve yet to boot them up (I neglected to pick up the third required battery), but I’ve taken them apart and started to scheme together an attack plan. I’d like to begin by dumping the ROM from a cartridge. I’m considering two primary approaches to this.
#1: Run code via the data cable
There may be a way to transfer a program via the data port to run from the device’s RAM. Should that be the case (…and we don’t run into any pesky security features), we can write, assemble, and transfer a program that dumps the ROM back via the data cable!
This is an undertaking, however – I’d need to obtain a data cable, make a board to plug it into a PC (though it’d be a simple one), and crucially figure out the memory map. And that’s assuming there is such a feature in the first place! I’d rather not fly blind, and given the lack of any development manual, I’ve put more work into thinking up approach #2, which is to…
#2: Tap into communications with the cartridge
As the machine accesses the data on the cartridge, we can tap into the signals it sends and receives and observe the communication patterns to deduce what purpose each pin might serve. Importantly, which pins constitute the address bus? And the data bus? How is a potential bank controller… controlled?
Earlier, I wrote:
I already expected that “only” to be overly optimistic, but y’all. Sling me a pack of paracetamol, because this’ll be an explosive headache. This is hardware in a different price class (and manufactured by a more hardware-focused company) than the Game Boy, where the bulk of my experience lies, and it shows: Already in 1987, Sharp was exercising some major bespoke minimization techniques. Far from the card-edge connector of most consoles, or even the card-edge-style connector found on, for example, the Master System’s Sega Card, or even even the funkier rendition on the PC Engine’s HuCard, Sharp’s solution is a 45-pin single-row pin connector with a 1-millimeter pitch.
Only 39 of the 45 positions are loaded, as evidenced by the jazzy bespoke device-side connector apparently manufactured by JAE, AKA Japan Aviation Electronics:
Who is she??? This is no standard connector (or perhaps it was in 1987 and is no longer). Thus, I’ll have to design a bespoke man-in-the-middle breakout board – one end plugs into the device; the cartridge plugs into the other; a third end exposes a standard (non-tiny) pin header. This won’t be entirely easy – for one, the Sharp cards are only 2 mm thick, so the board-plus-connector must be hwafer-thin, and for two, I’m not sure I can find a 1-mm-pitch male pin header that reaches deep enough into the cards – but… it certainly inhabits the realm of possibility. It’ll cost a bit, however, so we’ll see when I can comfortably set that cash aside.
And bleh! Epoxy! Nintendo never globbed any tops, and surely their games were produced in higher quantities than these. I suppose it makes sense with Sharp manufacturing their own chips that they might be able to glob the die on there for cheap even in smaller runs.
