Ethereum Nerd-Sniping: A Conversation With Veteran Dev Ben Edgington on the Merge and Boring Day Jobs
An update on the ongoing major upgrade to Ethereum with veteran developer Ben Edgington
As droves of executives on Wall Street and in corporate life have abandoned their safe roles for a gamble in the nascent blockchain and web3 space, a touch of a midlife crisis and boredom with one’s job is surely a driving factor.
Say you led the engineering and product management team within the information systems group for Hitachi in the UK, as Ben Edgington did for more than six years. He’d been with the Japanese electronics manufacturer for almost two decades by then when, in 2016, he discovered the Ethereum blockchain and a plan to change how the network achieves agreement called proof of stake.
“I was bored at work,” Edgington told me recently. “I was quite senior. I wore a suit every day. I spent a lot of time lunching, shaking hands with CTOs and CEOs, traveling the world – and it was fine, there was a certain attraction to it – but it wasn’t in my heart-of-hearts what I wanted to be doing.”
He would rather have still been getting his hands dirty in the technical details of projects, writing code, but he now had teams of people to do that for him, leaving him with meetings to attend, documents to review, hiring plans and budgets to write and other tasks he didn’t enjoy.
“I had the prospect of 10 more years of that until I can comfortably retire or doing something else, and this blockchain thing came up and just gripped me. Ethereum has this nerd-snipe thing, it just has this fascination” about it that makes some people become obsessed, he said.
The origin of nerd sniping from xkcd
The wider Ethereum community may want to tip their hat to Hitachi for its part in spurring Edgington to seek a new adventure. Since joining ConsenSys in 2017 he’s played an important role in moving the most-used blockchain toward the biggest evolution in its history. The switch to proof of stake from proof of work as Ethereum’s consensus mechanism has never been done to a live blockchain network, nor has anything like the scale with which Ethereum developers are approaching the switch been attempted.
So what is proof of stake? And for that matter, what’s proof of work? Put simply each is a way for a network to agree on its current state.
Bitcoin pioneered proof of work, which is used to add the latest batch of transactions to the blockchain history. In proof of work, computers all over the world race to be the first to solve a hashing function that can only be solved using trial and error. The winning computer, called a miner, is rewarded with free Bitcoin. Yet that means an enormous amount of computing power – and electricity – is required to confirm the latest Bitcoin block.
This energy consumption has drawn wide ranging criticism and is even possibly going to be banned in New York state. Since its inception in 2015 Ethereum has also operated under proof of work but is maybe only a few months away from switching to proof of stake.
“You cannot make proof of work environmentally friendly” — Ben Edgington
In that system, people put up a certain amount of the cryptocurrency ether to use as a type of collateral to secure the network. The crypto – rather than computing power – is the basis for the users of the network to do the right thing and confirm the latest block on the Ethereum network. If someone acts against the interest of the network, some of their ether can be taken away; conversely the people confirming transactions – called validators – are rewarded with free ether for their effort. In all, very little energy is expended in proof of stake, which will cut Ethereum’s power use by over 99 percent.
This switch to proof of stake is called the merge, and has been in the works for years. What’s amazing here is that when it comes time for the merge, Ethereum’s entire network will move at once from proof of work to proof of stake with – hopefully – not a single hiccup.
The task is also extremely complex because there are seven different software versions that allow network nodes to read Ethereum blocks and smart contracts. Some of them are Lighthouse, Nimbus and Lodestar. These seven software iterations are called clients and they help inspire network innovation and make Ethereum much more resilient to attack or buggy software.
Edgington is the product manager of one of these clients, called Teku, which has its roots in the protocol research and development group within ConsenSys. He’s also the author of the twice-monthly What’s New in Eth2 newsletter and the technical manual that will cover the process of why and how the merge was implemented.
“There are many incredible things about this, the way that we’re doing this with multiple client teams working semi-independently – having done so for the last three and a half or four years – to deliver” the project “all coming together separately to deliver this thing is kind of unprecedented,” he said. “It really is unique to Ethereum. It’s completely unprecedented in the blockchain world but in a broader sense is unusual. That speaks very strongly to the Ethereum ethos. It would have been a lot easier if we had one client.”
Ethereum inventor Vitalik Buterin was expecting the network to move to proof of stake as early as late 2017. It has not gone to plan, and there have been many versions of the network roadmap to move forward that have fallen by the wayside. For Edgington, who jumped into Ethereum because he thought the proof of stake design was so elegant, the wait is almost over. And he’s ready for the switch. He’s also motivated by environmental concerns to get Ethereum off proof of work.
“You cannot make proof of work environmentally friendly,” he told me. And rather than relying on a constant need to upgrade computer hardware to run a proof of work system, proof of stake internalizes the incentives of a blockchain using its native token ether to secure itself.
“Proof of stake is a proper crypto-economic mechanism,” he said.
The next test is when the Ropsten test network will be switched from proof of work to proof of stake. That’s slated for Wednesday June 8 and if it goes smoothly Edgington said he hopes there will be a bit of pressure to do the actual merge by August or so.
“I’m anxious and eager to get on with it,” he said. “I’ve been working toward this moment directly for the last three and a half years.” While it’s taken longer than anyone expected to get to this point, “We’re ready,” he said. “The time has come.”