Regarding proof-to-miner flaw: as devs
describe it's possible for the miners to get free trust. But imo it's possible modify this type of proof, giving money up to N miners.
The user sends a 0 output transaction with proof_money / N fee. The user waits N-2 blocks. Then s/he sends another transaction with N-2 outputs of proof_money / N, one for each miner that discovered a block; this second transaction has a proof_money / N fee.
Both transactions will have the GUID recorded.
In this way a miner must mine N-1 blocks in a row to get trust for free.
It's possible increase N in order to increase security, paying more overhead into the blockchain.
