Source: Deep Learning on Medium

Go to the profile of David T

I had concerns about how long it would take to modify the latest Bitcoin client so it would validate a Grassland node’s object tracklets alongside transactions and so its proof-of-work would instead be based on hashes of each node’s neural network’s computation, but thankfully because of intentionally similar protocol choices made early on and the fact Bitcoin’s fundamentals haven’t changed much over the years, placing a Bitcoin client “inside” a Grassland node has taken less time than I anticipated.
Of course, instead of pointing the whole thing at other Bitcoin nodes it’ll be pointed at other Grassland nodes who will not only be able to verify their “chain” of money moving from “wallet” to “wallet” but people and objects moving from position to position.
These protocol similarities will enable very interesting logic, such as letting you create active “bounties” as an extra payout to nodes gathering data from any locale you mark in the UI, ranging in size from half a centimetre on the earth’s surface to as big as a hemisphere.