ghash.io

Solving the Pool problem Currently the biggest threat to the viability of cryptocurrencies is the existence of mining pools. Pools take an inherently decentralised network and partially recentralise it for convenience. The most recent example would be the hullabaloo about ghash.io approaching 50% of bitcoin's hashrate. Pools are necessary because work rewards are too granular. You only get rewarded for contributing if you publish a block. Blocks cannot be fine-grained enough for effective solo mining, since they must remain synchronised over the network to prevent orphans and forks, and that takes time. What if the proof of work could be separated from the blocks? Blocks need to be carefully synchronised, but proofs of work do not need to be as up to date, since not every one is needed. I propose something like this: 1. A block is published, with a hash below some target to ensure the right number of blocks per hour are published. This part is identical to the existing system. 2. Some who receive this block start mining it again. They concatenate the block hash, their mining address, and a nonce, and when a solution is found, that is broadcast as well. However, these 'mini-proofs' would have much lower difficulty than the normal block hash, so anyone has a reasonable chance of finding one in an hour. 3. The mini-proofs are broadcast just like transactions, making their way across the network to those who are mining the next block the traditional way. 4. They are included in the next block just like transactions, and create tiny rewards to the payment addresses. The reward would be a fixed value, distributed evenly across all shares (or weighted according to difficulty, if different difficulties are permitted), so inflation is the same as normal block rewards. A few thoughts: The mini-proofs secure the network in the same way as the block proofs, but more granular. Instead of having to wait for the next block to see which branch has more hashrate behind it,