Translation Of Chinese Miner Consensus Meeting

Title: Support 2MB fork, reject any fork with <90% consensus

Author: LitecoinFarm – Jiang Zhuoer

On 23 January 2016, one of the coldest days in Beijing, HaoBTC organised a Bitcoin meeting to let participants discuss the Bitcoin network’s capacity, forks, mining, Bitcoin’s outlook, and more.

Regarding the block size increase, participants have made the following decisions and hope other Bitcoin enterprises and users will follow:

1: Increase the Bitcoin block size to 2 MB
2: Refuse to support any fork that has support of less than 90% of mining capacity

The 2MB block size originates from Satoshi Nakomoto’s whitepaper: “A P2P Electronic CASH System” instead of “A P2P Electronic SETTLEMENT system”. According to Satoshi’s design, Bitcoin is to be used as a transactional currency instead of for settlements. A 1MB block size limit does not satisfy the goals of a transactional  currency, and possibly not even a settlement one (note: 1MB ~= 7 tx/s, SWIFT ~=60 tx/s).

As for the 90% consensus requirement, the purpose is to ensure the Bitcoin community can develop in a healthy and sustainable manner.

Representatives from HaoBTC, Bitmain , Antpool, Antminer, BTC123, Yunbim Bither, Bikan, SosoBTC, and more.

More About The Consensus Mechanism

As we all know, the block size discussion is no longer purely a technical discussion, but has turned into a political issue.

During the Miami Bitcoin conference, there was no consensus among the community. To prevent the community from forking, we should try to lower our limit and accept others’ opinions, and at least achieve and some basic improvement. The consensus threshold should be high enough to prevent the community from breaking and to ensure future functionality.

Though the consensus we arrived at is not a big one; there are still plenty of people that hold the view of “1MB block size is sufficient” . We have made a small but great step towards raising the limit to 2MB. The problem with increasing the block size is that there are many conservatives believe risking a hard-fork would be very harmful to Bitcoin . Some rejected Classic not because of proposal to increase the block size, but because of the hard-fork risk, especially when only 70% network consensus is required. With the 90% agreement, we have made a great attempt to resolve the concerns of the conservatives.

Therefore the 2MB limit and 90% fork consensus should bring us to a more acceptable fork, and we wish to see consensus on other issues (e.g. increasing unconfirmed tx).

Under the 90% & 2MB consensus agreement, some have suggested the following:

1: Blocks are full
2: Core proposal is <2MB
3: Classic proposal have not gained consensus

Under the 90% hash power condition, switch from a 1MB limit to a 2MB limit to deal with the block size problem.

Segregated Witness
Core is attempting to use the Segregated Witness (Seg Wit) approach to solve the block size issue. While this has great potential, there are security concerns.

First, assuming Seg Wit is used by all one-signature tx, the effectiveness of it is only equivalent to a 1.6MB block size. Because of the complexity and development needed for Seg Wit, some may refuse to employ the update. Assuming 50% of tx use Seg Wit in 12 months, this is equivalent to a 1.3MB block size, which is pretty useless.

Second, Seg Wit requires a lot of manipulation of core components of Bitcoin, which carries a lot of risk. Bitmain CEO: “Currently Core devs are working overtime to catch up with the development schedule. We all know what will happen if we can’t get enough sleep and code on…”

Historically Bitcoin is far from “reliable” as we have all experienced. For instance, we had a ValueOverflow incident on 2010/08/15 in which billions of Bitcoin were generated in 1 single block. Once such issues occur, Bitcoin’s price will go down by a magnitude of 10 and destroy its reputation. Rejecting the Core immature Seg Wit proposal is very important.

Consensus action
We invite all Bitcoiners to join this basic consensus, please reply to this ( thread to show support of the above 90% hash power / 2MB block limit consensus.

Side Note : What Is 90% Hash Power Consensus?
As some hash power is anonymous, (solo mining/private pool), it is difficult to know their opinions. If we neglect the private pools’ share, we effectively need 95% of the public pools to agree.