Cute, but of course that approach will not stand a chance in any realistic application.

Just wait for the plastic circuits printers to become a commodity:

http://semiaccurate.com/2015/11/18/arm-charts-path-printed-p...

With this, a homebrew FPGA would become a reality. As well as homebrew ASICs in general.

What will be the technology size of these printers? I suspect the result will be circuits operating at a very low frequency compared to what we're used to.

Down to 50nm is possible. NMOS so far, so a bit slow, but I'd expect 1ghz to be achievable.