Déjà vu?
Although there isn’t a timeline yet, ethereum creator Vitalik Buterin today suggested that the platform’s developers may have to hard fork to once again solve pressing network problems. If pursued, the move would come less than three months after ethereum last executed a hard fork in a controversial bid to return funds to investors in The DAO.
Despite the potential risks (last time, the upgrade didn’t go as planned), the new fork seems to have initial support, as it is seen as a way to foil ongoing attacks on the network.
In recent weeks, the ethereum blockchain has been barraged with a series of denial of service attacks, which have slowed transactions and caused nodes to crash. Exchanges like Kraken and ShapeShift have temporarily paused ether trades as a result.
Because of this, developers are looking for a fast fix to the problem, which has acutely impacted clients including geth and Parity, in the form of a hard fork.
Ethcore developer Marek Kotewicz, who works on the Parity client, noted that, a fork increasingly seems like the easiest option to stop the attacks.
Kotewicz:
“You know, no one likes hard forks, but that’s what I think we need to do.”
In the last month, developers of the Go-version ethereum client, Geth, have released six new versions to fix the various denial of service attacks to no avail
But rather than continue to create new releases (the team has averaged one release every three days), the idea behind the hard fork is to resolve the network issues in one swoop.
Uncontroversial but controversial
But it’s important to note that this hard fork will likely be different than the one executed in the wake of the failure of The DAO.
As many ethereum developers see it, this hard fork is an uncontroversial change as it will boost security, unlike the last fork. Passed amid heightened drama and discussion, developers argue the most recent fork shifted consensus rules by altering ethereum’s immutable ledger, making it unclear whether its smart contracts could be truly “unstoppable”.
But since the next fork would bring a security change that developers expect to impact users positively, the feeling is ethereum’s community is likely to be less resistant.
“I don’t expect much controversy over this one, technical forks are basically just ‘updates’ for the protocol and this one is obviously necessary,” developer Zsolt Felföldi, who works on geth, commented.
Not to mention, hard forks are arguably part of ethereum’s DNA, as many in its community view the action as a way to upgrade the software in a way that supports its larger goals.
“The ethereum project always stated that it will hard fork often to progress the technology,” Heiko Hees, a developer working on ethereum’s Raiden Network, explained.
In the eyes of many ethereum developers, this hard fork will look more like the hard fork to update to Homestead, ethereum’s latest software release, which went according to a previously announced roadmap.
Felföldi said:
“As Homestead was easily accepted by everyone, I expect the same here.”
Why a hard fork?
The attacker has been taking advantage of the way the network prices a function called “gas”, which determines the fees used for accessing smart contract processing power on ethereum. Because prices are low, those familiar with the matter said the attacker can cheaply disrupt the network. This is what the developers are trying to fix.
So, while there are many versions of the hard fork in progress, the gist is the same. The fork will seek to raise gas prices to a cost that more accurately incentivizes actions that aren’t disruptive to the network. (Some multiply the cost by 10 to 20 times what it was before).
For example, the top version of the hard fork, EIP 150, addresses this.
In the social media post describing the proposed changes, Buterin explained that developers ultimately hoped to adopt a more dynamic gas fee market that can meet the changing demands of the network. But right now, it’s too complicated to roll out, and the team wants to push out a quick fix soon as possible to stop network disruptions.
“In the longer term, there certainly are ways that the gas schedule can be made much more flexible,” Buterin’s post explains.
Outside agreement
Even those in other blockchain communities have argued that a hard fork is the only option.
A developer for ethereum classic, the cryptocurrency that only exists because it continued the ethereum blockchain abandoned in the past hard fork, also argued in favor of the idea.
“With the upcoming Metropolis release, there will undoubtedly be more issues arising, so if they keep making small changes up to the metro release they run the risk of compounded attacks,” ethereum classic supporter Vic Arulchandran said.
He agreed a hard fork is the best option to fix the issues “ASAP”. (Notably, ethereum classic has been experiencing DoS attacks as well, and may have to consider similar measures).
Elsewhere, Blockstream principal architect Christopher Allen agreed that this is the most likely outcome.
“They’re actually going to have to do a hard fork, with all the risks of a hard fork, to fix their own weight calculation,” Allen said at last week’s Scaling Bitcoin conference.
However, some say that there are still risks involved.
Economist and investor Tuur Demeester argued that ethereum could find itself in a situation again where initial support for a proposed hard fork erodes due to unforeseen consequences.
“The hard fork that created ethereum classic caused a lot of animosity and a lot of dissatisfaction in the community,” he noted.
On the other hand, his criticism is not directed toward the specifics of the hard fork so much as what he sees as ethereum’s attitude toward security and how it has handled problems in the past.
“I would be concerned that it’s going to undermine the idea that ethereum is a stable network where people can actually deploy contracts that can sustain real-world business interactions,” Demeester said.
Summing up the likely views of many observers, he added:
“I’m curious to see how it plays out.”