Topic: Instead of 20mb blocks why not this solution? (Read 743 times)

Indeed, the most hash power ultimately does decide the longest chain, however the miners are ultimately at the mercy of the developers, full node consensus, merchants, and exchanges, and users.

There is a power dynamic between all of these groups. What happens when the community decides on something which upsets most of the developers (many of which voluntarily contribute to bitcoin core)? Do the miners and the community want them to walk away? What happens when a majority of the merchants and exchanges want something the miners don't, do the miners continue developing the longest chain of something that is as useless as most alts because no one accepts it and its just a speculative token?

Most of those Chinese miners probably are much better off with 1MB blocks because their poor internet connections but are willing to go along with a 10 MB limit because they need to consider the needs of the bitcoin ecosystem as a whole.

What is being tested here is if an open source project can find consensus without a "benevolent dictator" as Hearn wishes for bitcoin. In one sense he does make a good case as many open source projects do indeed function better with an ultimate decider like Linus with Linux, on another hand bitcoins core principles of decentralization make such a hierarchy somewhat contrary to the overall vision. 

No. Code changes are decided by the majority of Bitcoin blocks mined.  Nodes, retailers, and users will need to follow the majority of miners, or Bad Things happen.

Their opinion carries weight in the West at least. They'll also be in touch with the other big players in the infrastructure and perhaps form a united front when the time comes.

No one wants to be caught out on the wrong side of the decision.

No. Code changes are decided by the majority of Bitcoin Nodes. If one side decides to go to one code and the other don't, they will fork. Coinbase can't do anything except manipulating prices.

It could be done in such a way to maintain the social contract my adjusting the block reward to 1 minute target and 2.5 BTC reward.

This doesn't change the fact that this solution has been discussed many times before and isn't wise. Any alts with 1 minute less confirmation times have extremely high orphan rates and more forked branches. While there isn't an exact ideal number it certainly isn't 1 minute but somewhere between 2.5 to 20 minutes. Dropping it down to 2.5 minutes will increase the TPS from 3-7 to 12-28tps which isn't enough.

Even if we were ready for alternative solutions like the lightning network we would still need to raise the block limit regardless to handle that implementation.

A better idea is for us to form a consensus around some of Gavin's original calculations and go with a number like 8 MB block limit as it uses his own math, it won't be as drastic of an increase as Hearn is trying to push through with XT, and with 2 added qualifications:

1) We create a plan to address some of the concerns raised by those who oppose the increase, like having qualifiers such as creating a plan to increase the node count in a sustainable way and only roll out the code when certain goals are met, having a clear plan to test and roll out changes for the lightning network(fixing transaction malleability and starting to test it on a sidechain) and other proposals and meet some goals, and possibly finishing other privacy focused proposals like Todd's BIP 63.

2) We create perform more formal testing and simulation.

Hopefully, everyone can humble themselves and we can all work together on a solution which satisfies the community instead of being stubborn.

Changing the block timing and block reward structure radically breaks the social contract that Satoshi made when he created Bitcoin. None of the miners will agree to such radical changes and that will prevent them ever happening.

In case you missed the memo, the way Bitcoin works now is the code changes are decided by the big players like Coinbase.

1 minute blocks?
When a miner finds a block, the other miners spend time verifying the block that was discovered. If there were only 1 minute blocks the miner that found the previous block would have a large upper hand on the other miners since they don't have to validate their own solved block. Even if it only takes 5 seconds 5 out of 60 is almost a 10% loss in time that could be spent working on the new block.

Why not just make the block target times 1 minute by reducing the difficulty?

That way we could keep 1mb/.75kb blocks while speeding up the confirmation times. This will speed up the rate the transactions are cleared out and not have the orphan rate issue of the 20mb blocks.
All we would have to do is reduce the block reward to 1.25 btc by the halving.

Discuss.