Quote from: NewLiberty on October 16, 2014, 01:05:04 PM

Either we've gone through the looking glass, or else the goal is that Bitcoin should fail and some alt coin take its place?
Why hard fork Bitcoin to enable microtransactions if only to make them too expensive, as well as add risk and cost and also remove functionality in the process?

Gavin's 2nd proposal also seems worse than the first by the arbitrariness factor. x20 size first year, for years two through ten x1.4, then stop.

1) It's not about micro-transactions. It about the network having enough capacity to handle normal transaction (let's say over $0.01) as adoption grows.

2) It's not arbitrary. Gavin's revised proposal is better than the first because it is more finely tuned to match the current and projected technological considerations. Remember, the point of the proposal is not to maximize miner revenue or create artificial scarcity. It is allow the network to grow as fast as possible while still keeping full-node / solo mining ability withing the reach of the dedicated home user. That is not an economic question, but rather a technical one. Gavin and the rest of the core devs are computer experts and are as well equipped to make guesses about bandwidth and computer power growth as anyone.

Lets look at each of the three phases of Gavin's revised proposal. Step one: raise the MaxBlockSize to 20MB as soon as possible. That would be a maximum of 87GB per month of chain growth, or 1 TB / per year - easy to store on consumer equipment. Using myself as an example, I have a 20Mbps cable connection, which would actually be able to handle 1.4 GB every 10 minutes, so 20 MB blocks would utilize just 1/70th of my current bandwidth. I think most of us would agree 20MB blocks would not squeeze out interested individuals.

Phase 2 is 40% yearly growth of the MaxBlockSize. That seems entirely reasonable considering expected improvements in computers and bandwidth.

Phase 3 is stopping the pre-programmed growth after 20 years. This recognizes that nothing can grow forever at 40%, and that our ability predict the future diminishes the farther out we look. Also, lets image that in years 16 - 20, computation resources only grow by 30% per year, and the network becomes increasingly centralized. After year 20 network capacity would freeze, but not computer speed growth, so the forces of decentralization would have a chance to catch up.

Gavin - I hope you have enough support to implement this. You are the best chance for reaching consensus. And thanks for being willing to lobby on its behalf. 100% consensus is impossible, but 75% - 80% percent should be enough to safely move forward.

Everyone else - remember that without consensus we are stuck at 1 MB blocks. Gavin's proposal doesn't have to be perfect for it to still be vastly better than the status quo.