If this conjecture holds, or is a correct and useful implementation of the “function(claim or refund)” mechanism described in this paper, then the question arises whether a play would actually put up n-1 times the equity of his chips just to play a poker game.
But it seems to the author, at least upon a small amount of meditation that this amount of escrowed money could be put up by a 3rd party, that would basically act like an insurance company. At first this might seem absurd but its really not much different than the intended systems today.
Essentially we could think of such third parties as being somehow sort of independently contracted to provide trustworthy participants for the base of the poker games. ID systems of all sorts could be used for this.
If there is a basis for this it has obvious and significant implications and ramifications in regard to decentralized and ideal ID systems as well as collusion/reputation management in regard to Ideal Poker.
There is also relevance to drop-out tolerance here because players can also contract 3rd parties to provide their key’s in case the (honest!) player accidentally drops out.