The balance sheet is a smart contract abstraction of raw distributions from the sandbox.
Each item on the balance sheet is really a distribution of arrivals for some basket. This is literally true as each worker on the factory floor is swiping and tapping the flow. The assembly line prices the good from factory inventory to sales inventory, as does the stocking system at each end. Everywhere, a two color trade. Important to remember, worker have the red/green, so they know when their own flow is out of variance.
In business accounting we see a lot of flow ratios, sales to inventory and the like. All derived from the distributions in the pits and in the secure elements. Secure elements collect product ID, apps provide context and generate balance sheets.
Risk analysis therefore extends across balance sheets naturally, it measures risk of 'fails to deliver'
The sandbox is peculiar under the transaction costs zero, that is like absolute zero in physics. If card tapping is not a bother, the sandbox measures all sort of ad hocs, stuff humans are really great at. I seriously think the sandbox, this version anyway, will over use humans, find them more efficient then we really are.
Self discovery in this version is mainly the appearance of three color pits run by hedge funds, which are supported. They will come and go, keeping a check on the bot collective that it does not attempt a coup, a secret partnership with mad hatter tappers.
How to automate the hedge funds?
You need a Watson that does a continuing series of risk analysis on a some segment of smart contracts. It is looking to find conflicts and complements in the delivery mandates of the smart contracts. Then Watson would automatically open up a hedge fund with insurance promises. It offers savings, loans and insurance. Contract writers could fee the service out to Watson.
This works because Watson can see both the bit error and current pricing trajectories. It has a direct map between arrival uncertain and price via the container algebra. Smart contracts are difficult, and sometimes impossible, to price at the decision points. There is a good map for a Watson like device. The insurance model extends to market pricing of companies via balance sheet analysis, in sector fashion.
Coins are conserved within the bounded variance. A bankruptcy happens by stages, the number of basket sizes dropping, precision reported honestly. Hence, there are bounded exits points.
The consequences of this is that the three color hedges get killed, Watson reveals the discovery and the insurance payments get hedged elsewhere. This is a wash rinse and repeat, the boundary between singularity 1.2 and 2.0. It gives us the illusion that Watson is gravity.