[hiAndrewQuinn]

There's a paper from the 1990s of someone trying to do neural net-type things to construct an FPGA, and it gets really out there, like 'this gate isn't actually connected to anything but if we change anything about it the laws of electromagnetism cause the whole thing to stop working' weird. FPGAs are fascinating little dudes, esp in domains like HFT where their speed directly translates into dollars, but I totally believe you here.

[altermark]

https://www.damninteresting.com/on-the-origin-of-circuits/