Bitcoin Core developer Gavin Andresen today proposed a hard fork change for Bitcoin XT in order to allow for an increased block size limit on the Bitcoin network. So far, however, it has failed to appease most critics of his previous proposals to increase the block size limit.
It is widely agreed that at some point an increase of the block size limit will be needed to allow the Bitcoin network to handle more than seven transactions per second. The Bitcoin Core development team, however, has so far not been able to reach consensus on the correct timing and strategy to accomplish this.
Convinced that time for debate is running out, Andresen has therefore elected to shift his efforts to Mike Hearn's Bitcoin XT, a fork of Bitcoin Core. If Bitcoin XT overtakes Bitcoin Core, Andresen expects to be able to introduce bigger blocks on the network through this fork. He believes this would incentivise Bitcoin Core to adopt bigger blocks as well, or become irrelevant, as the two versions would become incompatible otherwise.
But while Andresen's proposal is being widely lauded by Bitcoin's Reddit community, it hasn’t appeased critics of his previous proposals. Most importantly, Andresen's decision to opt for a hard fork is regarded as dangerous by several prominent members of Bitcoin's technical community.
Bitcoin Core's lead developer Wladimir “wumpus” van der Laan has recently spoken out against a unilateral hard fork. He believes that it could risk a split of Bitcoin’s blockchain as a result of the incompatibility of Bitcoin Core with Bitcoin XT, and that it could lead to users losing money to double spends in the confusion.
Perhaps unsurprisingly, the new patch for Bitcoin XT has not swayed Van der Laan’s mind. While he did not elaborate on his concerns specifically when asked by CoinTelegraph, he did indicate that he broadly agreed with Bitcoin.org's recent policy update. The update reads:
“Contentious hard forks are bad for Bitcoin. At the very best, a contentious hard fork will leave people who chose the losing side of the fork feeling disenfranchised. At the very worst, it will make bitcoins permanently lose their value. In between are many possible outcomes, but none of them are good.”
This sentiment was echoed by Blockstream co-founder and hashcash inventor Dr. Adam Back, who has also been a prominent critic of Andresen's suggested Bitcoin XT fork. Speaking to CoinTelegraph, Back said:
“Everyone in the developer and technical community is strenuously advising to avoid the unnecessary risk of a unilateral hard fork. These are the very people with the knowledge and expertise we all rely on for the security and ongoing support and maintenance of Bitcoin today — and they are extremely concerned. The best approach for Bitcoin's continued success is if we work together in a calm, scientific, respectful and constructive way.”
Peter Todd, one of the first and perhaps most fierce opponents of past block size proposals by Andresen, agreed:
“I think this hard fork is a huge risk with the potential to both destroy Bitcoin in the medium to long term through centralization, as well as cause significant short-term harm to the network by the rushed and contentious deployment process he suggests.”
If Andresen does decide to move forward with the hard fork, it will take a while to deploy. First, the proposal will need to be accepted by Bitcoin XT lead developer Mike Hearn. This will probably not be a problem, however, as Hearn is a staunch supporter of a block size increase himself.
Bitcoin's block size limit on Bitcoin XT will then increase by January 2016 at the soonest. This will happen after a supermajority of miners run Bitcoin XT instead of Bitcoin Core, and explicitly agree on the block size increase. To be more precise, 750 of 1,000 consecutive mined blocks need to include a message in approval of the change.
Once an increase of the block size is adopted by the Bitcoin network, and after a grace period of presumably two weeks, the Bitcoin block size limit would first be increased to 8 MB. Effectively, this means that any Bitcoin node that still runs the “old” software at the time when a block bigger than 1 MB is mined would be divorced from the rest of the network. Furthermore, the maximum block size would automatically double every other year for 20 years, until it reaches 8 GB per block in 2036 — at the soonest.
Explaining these parameters on GitHub, Andresen wrote:
“The initial size of [8 MB] was chosen after testing the current reference implementation code with larger block sizes and receiving feedback from miners stuck behind bandwidth-constrained networks (in particular, Chinese miners behind the Great Firewall of China). The doubling interval was chosen based on long-term growth trends for CPU power, storage, and Internet bandwidth. The 20-year limit was chosen because exponential growth cannot continue forever.”
Apart from the suggested hard fork strategy, critics of Andresen's previous proposals are also not convinced this is the best way forward for Bitcoin itself. Speaking to CoinTelegraph, Todd explained:
“A genuine proposal would discuss how the change has been tested, and not just on a local machine, as well as what attributes of Bitcoin it is trying to preserve, and how it preserves them. What is the effect on profitability of small miners versus large miners, and under what conditions? What kind of margins are needed against DoS attacks? How do we pay for proof-of-work security against 51% attack in the long run if there isn't scarcity to drive transaction fees? Gavin has done nearly none of that kind of analysis, and the analysis he has done to date have been found to have significant flaws.”
Back emphasized that the question is not so much whether Bitcoin should, or should not, scale to allow for more than seven transactions per second, but rather how and when. He said:
“Everyone wants to scale Bitcoin, and there are multiple scale proposals under review. I think we will see a better outcome for Bitcoin’s safety and security if we continue the collaborative process that is evaluating and optimizing multiple scale related proposals at this time. Now that Gavin has a written proposal I am expecting it will be reviewed and evaluated against the others proposals that were published over the last month.”
CoinTelegraph reached out to Gavin Andresen and Mike Hearn, but received no response at time of publication.