As Censorship on Ethereum Begins, Could This Open-Sourced Code Help Counter It?
Flashbots, the team of developers behind MEV-Boost, a critical piece of software for the next phase of Ethereum, has decided to accelerate the open-sourcing of some of its code as the broader Ethereum community frets over looming risks of transaction censorship.
The move comes in reaction to the U.S. Treasury Department’s Office of Foreign Assets Control (OFAC) banning Americans from using Tornado Cash, a service that anonymizes transactions, because it abetted money laundering. It also followed the arrest of Alexey Pertsev, who wrote the code for the mixer, banning Americans from using Tornado Cash.
Read more: Are Crypto Mixers Legal?
MEV-Boost is an optional piece of software which, come the Ethereum Merge to proof-of-stake (PoS) in September, will separate “builders” who create blocks of transactions from the “proposers” who propagate those blocks out to the wider network.
Flashbots pitches its software as a way to help validators – the computers that process transactions on Ethereum’s PoS network – more easily (and equitably) extract Maximal Extractable Value (MEV), extra income that validators can earn by strategically selecting the transactions that they add to a given block.
But Flashbots sparked controversy last week after confirming that it would censor the default “relay” it uses to pass pre-built blocks from builders to proposers. In order to comply with OFAC sanctions, the Flashbots relay will exclude transactions involving addresses linked to Tornado Cash and other OFAC-sanctioned addresses.
Read more: Are Block Builders the Key to Solving Ethereum’s MEV Centralization Woes?
In response to community backlash, the team announced in a blog post that it would aim to get its relay code into developers’ hands earlier than planned. This will make it easier for third-party relays – rather than Flashbots’ own default relay – to launch with MEV-Boost this fall.
While Flashbots’ default relay will comply with OFAC sanctions, these third-party relays will be able to handle sanctioned addresses however they’d prefer.
The Ethereum community responds
How Ethereum developers and the community as a whole would handle network censorship was the elephant in the room during last week’s All Core Developers Meeting.
Indeed, it appears that some miners on the current PoW chain have already begun excluding non-compliant transactions.
Ethermine, the largest Ethereum miner, stopped including Tornado router transactions over a week ago pic.twitter.com/xw6R9u8mo3— takenstheorem (@takenstheorem) August 19, 2022
Micah Zoltu, the founder of Serv.eth Support, a support service specializing in Ethereum decentralized applications (dapps), led the topic of censorship on the call. In a conversation with CoinDesk, he remarked that the outcry around Flashbots’ decision to censor transactions was a wake up call for other relay providers. Now, Zoltu expects most relay providers to offer censorship-free relay options, and he thinks validators will prefer to run these non-censored relays.
Zoltu told CoinDesk that DeFi services firm bloXroute, for example, will run three relays: two that won’t censor, and one that will censor OFAC-sanctioned addresses. bloXroute will leave it up to validators to choose whether to use a censoring or a non-censoring relay.
On the developer call, many of Ethereum’s core engineers spoke out loudly against the movement towards transaction censorship.
Lukasz Rozmej mentioned during the call that if Ethereum creates code that would allow for censorship, that would make Ethereum developers the “enforcers” of censorship on the protocol, a role which developers would not likely want to take on.
Dankrad Feist, a researcher at the Ethereum Foundation, argued on the call that Ethereum developers should be adamant about censorship resistant qualities.
No one, however, had a clear plan for what developers should do to combat censorship ahead of the Merge.
If anything was clear at the end of last week’s call, it was that this is only the beginning of a long debate.