Sunday, April 23, 2017

The new probabilistic grammar coming from smart contracts

I have the general scheme of smart contracts as computing fails to deliver, a convolution between pure cash 'events' and the Euler surfaces of contracts.  We get a more detailed scale of contract failure, and pay offs thus more accurately scaled.

But the evolved grammar will look identical to existing contract law, except all the redundant terms r=moved, and all nodes have the grammar of  fails to deliver.  We quickly end up with a contract language operationally understood by the AI and humans alike.  It will cause the web to change slightly, all the semantic nodes have a slight pricing surface, so that, with a few bits of precision a semantic match can be had.  At that point we get true knowledge discovery, we and the web agree on language.

Fails to delivery should be non-stationary, the odds change as bets pile in.  Contracts have to specify a reference point, (annual measure, last week, etc).

No comments: