This is the seventh in a series analyzing the pettycoin implementation against Gregory Maxwell’s writeup on scaling. The first talked about UTXO commitments vs backrefs, the second talked about Propogation servers vs prev_txhashes, the third talked about Hashed Fees vs Random Extrapolation, the fourth talked about Simplified Transactions vs Normal Bitcoin Transactions, the fifth discussed Fast blocks vs Bitcoin blocks. and the sixth Fees and Horizons for A Microtransaction Sidechain.
Lacking the ability to mint coins, rewarding miners for a sidechain is difficult. Relying on fees in the initial stages is not possible, and using fees provides a disincentive for actually using the network.
To solve this problem, pettycoin allows transactions to flag themselves as feeless, with the understanding that miners may not include such transactions. It also mines the future to pay for the present: 25% of the fees from block N+(X-years) would get paid to block N. The actual number of years was to be selected by an on-blockchain election conducted at the 1 year mark.
The problem with this is that it requires the retention of all old block headers, even before the horizon (otherwise we could use compact SPV proofs for those headers). A further lottery process could select an agreed subset, however.
A similar idea was suggested in the sidechains whitepaper, via a different mechanism: that old miners would have the right to mine a block with lower difficulty in the future. In some ways this is nicer, since the easy miner could provide proof of their old block using a compact SPV proof. Also, if the ability to mine an easy block is part of the UTXO set, UTXO commitments can provide protection against double-spends.
This introduces the issue that some party could flood the chain with easy blocks; perhaps instead each block would only be (say) 25% easier than normal, rather than ridiculously easy. The miner could then sell the use of their private key.
Weak blocks make checking significantly more complex, but I’d have to implement it to really know how bad it was. It fits fairly nicely into the bitcoin model of inputs and outputs, however.