Last week I summarized the contentious issues that have caused the price of bitcoin to fluctuate with one camp calling for a bugfix called SegWit that just happens to increase capacity (the Bitcoin-core developers) and the other camp that wants a quick fix to double the size of the blocks (the Chinese miners) in what is not really about network capacity but battle over control over the world’s most valuable cryptocurrency.
If the Chinese miners do pull off their threats to do a hard fork, what will happen? It is a fork, meaning that the original chain could continue alongside the big 2MB-block chains. A similar situation happened back in July last year when Ethereum split over the issue of the DAO resulting in the original chain continuing as Ethereum Classic (ETC) and the new chain that returned the stolen Ether cryptocurrency as Ethereum (ETH), rather confusingly as it is the new name that abides by the old rules.
That is possible in the world of Bitcoin too if the miners call their bluff and indeed, many exchanges have had to put forward statements on which currency they will consider as the user of the original BTC ticker. Indeed one company, First Bitcoin Capital, has even created a derivative coin called Bitcoin Unlimited Futures that is traded for investors to gamble on how the forking saga will end.
Some say that in a way it is as if they are doubling the amount of Bitcoin from 21 million to 42 million. In a way. The reality is that if the miners get their way it will mean that they can do pretty much whatever they want at a later date which does not bode well for the decentralized future of the cryptocurrency.
However, it is not that simple and much was learned from the Ethereum and Ethereum Classic fork.
Even if the two sides agree to disagree and to go their separate ways with one side going SegwWit on the original chain and another going with Bitcoin Unlimited, there is the issue of the replay attack. Since each of the existing 16 or so million bitcoins already in circulation will be duplicated, that means that the bitcoin addresses will be duplicated too. That also means that after the fork happens the networks will be open to replay attacks where someone can take a payment in one network and replay it in the other as both will use the same addresses and same keys to sign those transactions.
The BU miners say that with their superior mining power they can “solve” the problem by producing empty blocks for the original bitcoin network and effectively prevent transactions there from ever being mined into the Blockchain. Solve that is, by killing the original chain with their superior mining power. This threat has not gone down well among Bitcoin-core supporters, to say the least.
However, late last week a new twist emerged in the saga – a patented algorithm called ASICBoost.
A simple way to describe ASICBoost is a redesign of the hashing or mining algorithm used in Bitcoin mining by sorting part of the headers so that part of the calculations can be reused from block to block. The number of calculations can thus be reduced by about 20%. Or in other words, by pre-sorting the candidates to mine, they can increase mining output by about 20%.
However, there are two issues with ASICBoost. First off, the blocks must need to be somewhat less than full – and indeed we have seen many Chinese miners mine empty blocks and forgo the 2 BTC mining fee – and secondly ASICBoost is incompatible with SegWit.
The world’s largest Bitcoin miner manufacturer Bitmain, whose CEO Jihan Wu is one of the most vocal opponents of SegWit, claims a patent on ASICBoost. However, he was quick to point out that Bitmain has not used ASICBoost on mainnet (the main Bitcoin network, not the test network or testnet).
Whether others using Bitmain equipment have done so or not is unclear but the existence of empty blocks would suggest that some are at least experimenting.
The Bitcoin Unlimited group are trying to push a different technology called Extension Blocks instead of SegWit to achieve much of what SegWit can achieve (fix the transaction malleability bug and enable Lightning-like networks). However, it has been pointed out that SegWit had to be tested for almost a year before they felt it was ready for production and now the Bitcoin Unlimited supporters are trying to push for something new and untested that might not even work just to avoid SegWit.
Circumstantial evidence, surely, but suddenly all the chaos of the past two weeks is starting to make sense. It is not so much a battle for control by irrational Chinese miners who are risking everything – including the future of cryptocurrency itself – to wrestle control over the western developers. Rather it is about a small group of economically rational miners simply trying to preserve a 20% cost advantage without revealing its nature to others.
A number of high profile Bitcoin Unlimited supporters have jumped ship after ASICBoost was exposed. A major funding drive to increase support for Bitcoin Unlimited just failed and the price of Bitcoin, at the time of writing, has increased and is back over $1,200 again, not quite but almost at pre-SegWit drama levels.
The joke of the day that is making its way around the industry is that there are two types of Bitcoin users in the world – those who understand SegWit and those who support Bitcoin Unlimited. With these further revelations and understandings, it has never been more true.