I did something similar with electronics. I thought if I applied voltages to random pins, and read voltages on other pins, I could figure out how an integrated circuit worked.

The crazy part is that my strategy actually worked on a single MOSFET transistor, which only has three pins. I just assumed it would scale up on computer chip with tens to hundreds of thousands of MOSFETS and a few dozen pins. Also, if my first test had been on a BJT transistor, instead of a MOSFET, I would have destroyed it.