The coming bitcoin fork

Monday 12 September 2016

With the arrival of SegWit, it seems inevitable that bitcoin will fork, with Ethereum offering a model for how it might happen.

SegWit is a long-overdue and much needed addition to the bitcoin protocol. It will solve the malleability bug that offered such a convenient fig leaf to MtGox. More importantly, it will raise the limit on transaction volumes by allowing more to be fit in the same block, thereby avoiding the Elvis Scenario

The ideal is that SegWit will be adopted near-unanimously. The activation threshold is 95% of hashrate - a truly ambitious proportion, but one that would ensure the vast majority of miners came on board. It’s also a soft fork, which means updating isn’t - at this point - strictly necessary. Nodes without the SegWit code won’t recognise SegWit transactions, but they won’t be locked out of the network entirely. It's actually a pretty nice fix.

Potatoes fork

Forks can be dangerous

Not like Ethereum

This is one reason that the coming bitcoin fork isn’t quite the same as the one that just split Ethereum in two. Ethereum, if you remember, decided to hardfork to address the theft of funds from the flagship DAO project. A hacker found an exploit in the 'smart' (ahem) contracts code and drained the equivalent of tens of millions of dollars in Ether into a new address.

Indications were that Ethereum’s miners were on board with the idea of the hardfork to the tune of 95% of hashrate - the same as the activation threshold for SegWit, which is an interesting coincidence. 95% seems like a landslide, but what happened next suggests that the dynamics aren’t that simple.

In this article I have argued that ideology wasn’t the most significant factor in the split of Ethereum to ETH and ETC: it was money. Hashrate suggests that miners decided to start mining the more lucrative Classic, until equilibrium was reached. This site shows that hashrate and price still correlate closely: miners go for whichever chain is the better deal.

This raises a question over what will happen to bitcoin once SegWit goes live. Probably not much, at first. The two camps can still play together. But it marks the start of a divergence, those who support SegWit from those who, for whatever reason, do not. Once started, that divergence will inevitably grow; it can’t do otherwise. And there will presumably come a point where that divergence is formalised with a hardfork, albeit some way down the line. Many prominent developers and mining pools have called for such a solution.

To date, bitcoin’s hard forks have only ever been done for security reasons - they are not what anyone could call optional. As soon as one is carried out on the grounds of ideology, or anything else, there will be dissenters. That, too, is inevitable in a decentralised system.

Ethereum is an interesting case study of what happens when that occurs. Where there are two chains, miners follow the money. Even 95% consensus can’t stop that.


comments powered by Disqus