User blog comment:Pellucidar12/Proof length function/@comment-24920136-20161228041401/@comment-30754445-20161229000016

Well, there's also Friedmann's description of a "transcendental integer" which refers to numbers provable in 2^^1000 symbols.

My point was that there seems to be a consensus that a modest tetrational number is enough for this kind of things.