The post addresses this very issue:

> Precisely because the Turing machine model is so ancient and fixed, whatever emergent behavior we find in the Busy Beaver game, there can be no suspicion that we “cheated” by changing the model until we got the results we wanted.”

Following BLC8's bytewise encoding convention of [1], w218's binary encoding 0100 0101 1010 1000 0110 0110 0000 0001 0101 1011 1011 0000 0011 1001 1101 0 gets padded with 3 arbitrary least significant bits, say 000, and becomes 45A8_6601_5BB0_39C0 in hexadecimal.

[1] https://www.ioccc.org/2012/tromp/

> Precisely because the Turing machine model is so ancient and fixed, whatever emergent behavior we find in the Busy Beaver game, there can be no suspicion that we “cheated” by changing the model until we got the results we wanted.

Sorry; no winning for cheaters:-(

To be pedantic, that is a instance of the Berry paradox [1] and no you can not [2] as that would be a violation of Godel's incompleteness theorems.

edit: To clarify further, you could create a new formal language L+ that axiomatically defines 0 as "largest number according to L", but that would no longer be L, it would be L+. For any given language with rules at this level of power you could not make that statement without creating a new language with even more powerful rules i.e. each specific set of rules is capped, you need to add more rules to increase that cap, but that is a different language.

[1] https://en.wikipedia.org/wiki/Berry_paradox

[2] https://terrytao.wordpress.com/2010/11/02/the-no-self-defeat...