User blog comment:Wythagoras/All my stuff/@comment-10429372-20130718073726/@comment-7484840-20130718172023

My point was that the rule about not using the ¥ function could be easily bypassed by a function with a small, trivial difference, like being the largest number that can be defined in 2n+3 characters of english, rather than just n. the only real difference this makes is that the z function is not the ¥ function, and so could be used by the ¥ function.

I meant that the output of ¥ is the largest possible number defined in ...etc., not that the ¥ function was the output of a function.

Of course ¥(n) could be a definite function, and indeed probably should be. The problem is there is no limit to how large these functions can be.