Here is the original article the Loadstar referred me to. It is a picture a bit too rosy, perhaps, but quite clear.
It doesn’t say that blockchain’s smart contracts will turn logistics people from a handful of paper shufflers to a handful of logicians whose job is to read and prove out the code of smart contracts. That code will be embedded and executed in a transaction, without oversight, so it had better be right. How do we prove that? The arguments will shift from arguing about how the actual contract matches with what happens, and goes wrong– to arguing about both that, and whether the smart contract program had a bug in it.
I think there’s a new frontier for provability software, that can read a program and match it with some specs to see if it matches. The computer mathematicians already have software that can inspect theorem proofs and say if they are valid or not. I think we need similar software.
Remember every program has bugs.