After Ethereum went live on July 30, 2015, Vitalik flew to Asia. He visited Korea and then China again before heading back to Switzerland for a meeting to introduce the new board members and Ming Chan to each other. Even before the meeting, though, there was trouble among the newest additions to the Ethereum Foundation.
Ming had been in touch with Vitalik when he was in Asia to tell him that she and Wayne Hennessey-Barrett weren't getting along. He'd yelled at her for two hours, she told Vitalik.
“She was very unhappy with how they treated her,” Vitalik said. Then things took a turn for the worse after Ming sent a text message to the wrong person.
“The consternation was in the open because at one point, Ming made this mistake where she was complaining about the board, and was talking to her sister – well, she thought she was talking to her sister about the possibility of finding a different board,” Vitalik said. “Instead of sending some text messages to her sister, she sent them to one of the board members.”
With that setup, the new Ethereum Foundation board members and executive director were about to meet in person for the first time. Ming arrived a day early and Vitalik tried to talk her through how to get past what had already happened. Board members Lars Klawitter and Vadim Levitin had made the trip to Zug for the meeting, but after getting in a row with Ming, Wayne Hennessy-Barrett didn't show up. He participated by video conference.
Ming wore a suit for the meeting and ran it very professionally with timed topics – 5 minutes on this issue, 15 minutes on that issue – all in the hopes of avoiding any discussion of how everyone already hated each other.
Her tactic worked. But then Lars and Vadim left and immediately went back to their hotels. This didn't sit well with Vitalik. On the one hand it showed him how uncomfortable the members of the new management team were around each other. And it also showed him that there wasn't much passion for Ethereum in this new group.
“They weren't really that deeply interested in us or Ethereum,” Vitalik said. “They just saw it as a job.”
The meeting spanned two days, and on the second day Vitalik held a private meeting with the three board members. “The board members were, like, ‘hey, dude, we've got to get rid of Ming.’” Vitalik agreed, and not even a month into her tenure, Ming was told she wouldn't be continuing as executive director.
The next meeting that day was taking place at the offices of MME, the foundation's lawyers, where they were going to discuss Swiss regulatory issues. Ming cried as she walked there with Vitalik, Lars, and Vadim. “Ming wasn't invited to the lawyer meetings we were going to because she was now not going to be executive director,” Vitalik said.
Yet while he was in the meetings with the lawyers, Vitalik again got the impression that Lars, Wayne, and Vadim just weren't really that into them.
“I remember during the meeting, looking at the other board members' reaction, the way that they were interpreting the information,” Vitalik said. “I expected that I would feel warm and safe because there were these high-powered, professional people with lots of experience helping us now. But their body language and mood was like ugh, huh, okay okay. They were just so clearly not into it. That's when I realized I was not comfortable with them being the board.”
After months of searching and a public call for submissions for the board and executive director positions, after in-person interviews and negotiations, the Ethereum Foundation professional board – the one that would take the group to the next polished level – had been blown apart in less than two days. This wasn't another bump in the Ethereum road. This was the road.
A week later, Vitalik was in Toronto. Ming had stayed in touch even though she'd been told her services were no longer required. But Vitalik grew to appreciate her dogged interest in the foundation. This was the sort of passion and dedication to Ethereum and the foundation Vitalik wanted to see.
They began working together in Toronto on foundation issues, like an upcoming financial audit. None of the other board members knew Ming was still in the picture, and certainly not that she was now in Toronto working daily with Vitalik. On September 1, 2015, Vitalik sent an email to the board members updating them on his thinking about the foundation management.
He emphasized that he didn't want to do anything rash, but that big decisions needed to be made. “I feel that I, perhaps, in a lot of ways, have been too hasty in agreeing to things both there and many times previously in my career,” he wrote. He said any decisions would be postponed until September 10, after he'd sought advice from others about what to do.
Ming, he noted, had been doing good work on the upcoming audit. “I got quite scared when seeing just how much work there is to be done and realizing that we don't have any dedicated staff to do it,” he wrote. “She would be extremely hard to replace if she were to stop working tomorrow, and the current ongoing audits are a critical matter that would severely impact the foundations' situation if we were to miss deadlines.”
Even though she had been told that she'd been let go, Ming was on a contract as the executive director of the foundation, and the board members had been pushing Vitalik to finalize her removal. But he put that off. When he shared the September 1 email with me, he described it as “when I was bullshitting to procrastinate for a week about firing Ming.”
A week later, he sent an email to the three board members and the foundation's lawyers, saying he wanted the board to step down. They agreed. Ming now held the executive director position, as well as a board seat, and would come to exert great control over the foundation in the months to come.
A graduate of MIT, where she'd studied computer science and media arts and sciences, Ming had been born in Switzerland to Chinese parents. Her family moved to Michigan, and Ming had worked for the University of Michigan from 2007 to 2015, where she helped the school maintain web content dedicated to Chinese art, history, and culture. Her first task as head of the Ethereum Foundation staff was to cut costs, which were out of control.
In 2015, the peak monthly outlay for the foundation was over 400,000 euros. That covered the cost of developing Ethereum on the Go and C++ clients, and included money spent on hosting web sites, maintaining existing servers, and covering other information technology expenses. One of the larger chunks went to paying executive and staff salaries. They were bleeding money and weren't helped at all by the fact that half of the capital they raised in the crowdsale – $9 million – had been wiped out when the foundation didn't hedge its exposure to Bitcoin in late 2014.
“Ming turned all her attention to cost cutting,” Vitalik said. “At the time, we were basically completely out of Bitcoin and completely out of fiat [currency]. All that we had was our three million ether. We had to start selling huge amounts of ether to survive.” The three million ether the foundation had left was down from the six million it had been allotted from the crowdsale because three million had been allocated to the developer purchase program and all of it had been bought up by devs. At the time of the Ethereum launch in July 2015, the foundation had $1 million in cash. By mid-September all that was gone.
Vitalik loves precise numbers, and as we spoke about the financial straits the foundation found itself in in late 2015, he said he was able to arrange a private ether sale to raise funds. How much ether? “I managed to sell 416,667 ether for $500,000,” he said. The buyer was a firm working with Bo Shen, whom Vitalik had been courting in China for many years. “So, we were happy.”
The lack of funds had repercussions in the world outside the Ethereum Foundation, as their 2015 annual developer conference, DevCon 1, was set to take place in a few weeks in London. Yet the money wasn't there to put on the show, and the foundation considered canceling the event. When Joe Lubin got wind of this – Joe Lubin, who had now hired staff at his Ethereum application incubator ConsenSys – he said no way.
“Then Joe stepped in and said, basically, ‘no, we're doing DevCon 1 and if not, ConsenSys will organize a conference,’” Vitalik said.
None of these negotiations were known to the outside world. If all you had for your information about Ethereum was the blog on their web site, you would have been hard pressed to know what was going on. A post on September 2, 2015, announced that DevCon 1 – set to begin in a month – had been postponed until further notice. The foundation claimed the postponement was due to not being able to secure the London location they wanted on the dates they had scheduled. Behind the scenes, the situation was far more tenuous, and Joe Lubin and the people he'd hired to build the for-profit ConsenSys were about to run headlong into Ming Chan's new and improved nonprofit Ethereum Foundation.
●●●
The first Ethereum meetup in New York City took place two weeks before the ether crowdsale in the summer of 2014 at the Cardozo law school, spitting distance from Union Square. Hosted by Aaron Wright, an associate clinical professor of law who specialized in Internet and new technology issues, one attendee at the meeting had come from the world of medical database companies, the firms that process insurance claims electronically. His name was Andrew Keys and he would become one of the first employees at ConsenSys.
Through the business of his company, MedAZ, Andrew had come to understand everything that was wrong with databases and with payments. Nothing connected to anything else, and it took days or weeks to process some claims. Errors were rampant. He wanted a way to bring payments and database entry together in one holistic system. Bitcoin had done that to a degree, but Andrew soon realized its limited capacity to handle complicated transactions. Then he discovered Ethereum.
“I followed the blockchain ecosystem but didn't get professionally involved until I started seeing the promise of smart contracts,” Keys said. Once he'd seen the light, Keys saw the notice about the Ethereum meetup at the Cardozo law school and went with the intention of meeting Joe Lubin.
“I remember Joe being so confident that this would change how society works, I just fell in love with that notion,” Andrew said. “Once I started to understand the implications, I couldn't stop thinking about it. I still can't stop thinking about it.”
An Ethereum blockchain application that really drove it home for him was the update on what could be done with double-entry bookkeeping. That's the system – developed by a monk in 1494 – where one party to a transaction puts into their books that they were paid 10 dollars and the other party to that transactions puts in their books that they paid 10 dollars. But what prevents the side being paid from claiming they received 5 dollars and the side doing the paying that they forked over 15 dollars? They could pocket the difference. With Ethereum, a new type of accounting – triple-entry bookkeeping – became possible.
“There was nothing that basically watermarked that transaction, there was no handshake,” Keys said. With an unchangeable ledger in the middle of that transaction on Ethereum, there would now be a receipt. “That basically automates auditing,” he said. “KPMG, Ernst & Young, Deloitte, PWC, how they work, will be radically transformed.”
He was willing to put his money where his mouth was and offered to work for Joe at ConsenSys for free, which he did for six months between March and September of 2015. Lubin had never let go of the idea of Ethereum as a for-profit enterprise, even as the cofounders voted against it in the Zug meeting in June 2014. Now Joe had his own for-profit design studio, and he was pushing everyone involved with ConsenSys to make sure the commercial side of the Ethereum project had all the funding, advice, and support it needed to create the applications that would make or break the underlying blockchain ecosystem he wanted to create.
That meant, in the bigger picture, that as the Ethereum Foundation was running out of money in the fall of 2015 and telling its community that DevCon 1 was on hold, it was all systems go for ConsenSys to seize the opportunity to start driving the conversation in the direction it wanted. Throwing in a bit of cash and lining up big name sponsors wasn't a bad way to accomplish that goal.
“I did loan the Ethereum Foundation something like $100,000 to put on DevCon 1,” Andrew Keys said. He then sent an email out of the blue to an executive at Microsoft who happened to be playing around with Ethereum. His name was Marley Gray, and Andrew decided to reach out to him based only on the fact that Gray had the word blockchain in his LinkedIn bio. They soon hit it off. Andrew is nothing if not persuasive and can talk the balls off a rhinoceros. Andrew and Marley began collaborating on a partnership. Andrew also pitched the idea to Marley that Microsoft should be a lead sponsor of DevCon 1. Whether the event would be a ConsenSys undertaking or whether the Ethereum Foundation would get its act together in time to put it back on was still up in the air.
For Joe, the stakes were too high to not have a developer conference. “I felt like we needed an event to catalyze the ecosystem,” Joe said. “I suggested we create it and if the foundation will put its name on it, we'd call it the Ethereum Foundation DevCon.”
Then in late September, Vitalik announced that DevCon 1 was back and would take place at Gibson Hall in London, on November 9–13. In October, the foundation said that Nick Szabo, the inventor of the term smart contracts and the idea behind them, would be the keynote speaker. The Ethereum Foundation branding was starting to take shape as well. In that October release, Szabo was quoted as saying that Bitcoin was like a pocket calculator while Ethereum was like a general-purpose computer. I'd first heard this comparison made by Joe Lubin, though a bit later in 2016. Szabo's initial work with smart contracts was an attempt to allow contract law to be embedded into the programming languages that make the Internet possible. At the end of the release, the foundation tried out a bold new marketing ploy: “Ethereum is how the Internet was supposed to work.”
The collaboration between Andrew Keys and Marley Gray paid off in November, when Microsoft announced it was adding Ethereum to the applications available to its customers through its cloud services unit, Azure. They called it Ethereum Blockchain as a Service, because this was a version of the blockchain meant to be used by banks or companies that wanted to experiment with it to see if it could streamline their operations. This enterprise version of Ethereum had many of the public features of the blockchain removed; for example, there were no miners needed to verify transactions and the participants in a private network would all be known to each other.
Marley Gray was a big fan of Ethereum from early on. In the announcement about the deal with ConsenSys he wrote, “Ethereum provides the flexibility and extensibility many of our customers were looking for. With the Frontier release last summer, Ethereum is real and has a vibrant community of developers, enthusiasts and businesses participating.”
The partnership, importantly, was between Microsoft and ConsenSys, not Microsoft and the Ethereum Foundation. This was also announced on the first day of DevCon 1. It may have been the first shot across the bow of the Ethereum Foundation ship, which Ming Chan was now steering.
Andrew Keys said that ConsenSys at the time was the only outfit in the Ethereum world building anything for the blockchain's application layer. This led to ConsenSys dominating many of the DevCon 1 talks, “And that really pissed Ming off,” he said.
Joe Lubin said that initially he and Ming got along well, but that their relationship soon soured.
“She demonized a lot of people and she demonized me the most because we were an evil commercial force in her pure, nonprofit world,” Joe said. “Ming is a complicated individual. Very capable in many ways, but prone to extreme emotional situations which many, many people witnessed many times.”
I tried to get Ming to talk to me over the course of several months in 2019 and 2020, but she refused. From what I've been told by others, Ming felt as strongly about Joe as he said she did about him. There was no love lost between her and ConsenSys.
Andrew Keys didn't see eye to eye with Ming, and said she made several business decisions that hurt early Ethereum adoption. A big one involved IBM, which was considering using an altered version of Ethereum for its blockchain research and development. The deal would be enormous for the fledgling foundation. “IBM has a tremendously powerful distribution arm,” Keys said. “I didn't appreciate until ConsenSys how embedded IBM is into Earth – all the central banks, all the banks, all the supply chains.” Keys said Ming wouldn't take calls from IBM executives Jerry Cuomo, vice president of blockchain technologies, and John Wolpert, a global product executive for blockchain. IBM ended up creating its own blockchain, Fabric, for its R&D.
The loss of the IBM deal created a huge schism amid the broader community, Keys said. “You could have had everybody working on a private Ethereum,” he said. Ming wasn't the only one in charge at the foundation, of course, but Keys still thinks she bears a lot of the blame. “I lay a lot of it on her doorstep,” Keys said.
Another business decision that came to hurt Ethereum had to do with how it went about choosing its open-source code licensing. This may have been an extension of Vitalik and many other early founders' allegiance to being nonprofit and ensuring that all of its code was open source, or available to the public for free. That's all fine for the main version of Ethereum that went live in July 2015. But if a corporation wants to use a private version of Ethereum to build applications for its business, it wants to be able to keep the proprietary rights to the code it created. Two of the main types of licensing are GPL (GNU General Public License) and Apache 2 (Apache License 2.0).
GPL doesn't allow for the code that's written atop its system to be commercialized; its rights are only granted if the derivative product is also made free to the public. Apache 2 licensing is looser in this regard and allows commercial use by the application's creators. All of the clients built by Ethereum teams – Go, C++, Python, and others – had been built using GPL.
“Enterprises are allergic to GPL,” Keys said. “Any attorney for any company, when they see it's Apache 2, they'll say yes to the software, if it's GPL they'll say no.”
To those inside of ConsenSys, the Ethereum Foundation wasn't quite ready for prime time. But it was still so early in Ethereum's development that there was no comparable force like the foundation. In time, not doing anything about this would turn out to be something Ethereum could no longer afford to do.
●●●
When Marley Gray began tinkering with Ethereum in August of 2015 at Microsoft's innovation lab in Manhattan, no one around him had any idea what he was up to.
What he really wanted to do was to get nodes on a private Ethereum network to talk to each other, but he was having a devil of a time getting it to work. He didn't know how to write code in Go, the client he was experimenting with, and there was no one at Microsoft he could go to for help. Then Andrew Keys reached out and Marley was amazed that ConsenSys was just over the East River in Brooklyn. He went to their offices in that graffiti-covered building in the Bushwick neighborhood, when the furnishings consisted of picnic tables and there were 10 people there, tops.
“Joe would sit in the corner, and we just pulled up chairs and I told him what I was doing,” Gray said. He'd spent an entire weekend with very little sleep trying to get two Ethereum nodes to talk to each other, with no luck. He knew businesses and banks wouldn't have that sort of patience, so if Ethereum wanted to appeal to them, they'd have to make the tools available much easier to use to set up a test Ethereum environment.
ConsenSys wrote some code for use with Linux and Marley handled the cloud computing side and soon they had the Ethereum Blockchain as a Service product ready. Marley specialized in financial services innovation for Microsoft, so he knew that Morgan Stanley and Goldman Sachs wanted to experiment with private blockchains. The demand was there.
The time was right at Microsoft to be taking gambles on innovations like Ethereum. Satya Nadella was CEO at the time, and he'd sent word out to every Microsoft employee about his desire for them to explore new ideas and have what he termed “a growth mind-set.”
“I said this is a growth area, I guarantee you,” Marley said. After some haggling and a positive story in the Wall Street Journal about Microsoft's work with Ethereum, Marley got the okay from his bosses to be a lead sponsor of DevCon 1 and to also send him there to give a presentation.
Gray and Microsoft got a frosty reception from the Ethereum developers in London. “It was a lot of, ‘what are you doing here?’” he said. “It was not very welcoming.” In his presentation he'd tried to emphasize that Microsoft was now in the cloud computing business and that they didn't want to create their own blockchain. Microsoft wanted to make it easier to use Ethereum by providing cloud-based tools to Ethereum developers. He thought his talk went over well and waited to see how the press portrayed it.
By the next day, he hadn't seen a single story about Microsoft's interest in Ethereum or about anything he'd said. Two friends had joined him in London, and they were eating lunch at the restaurant at Level39, a tech incubator and shared working space in the heart of Canary Wharf. Marley was frustrated that he hadn't made an impact and that he couldn't seem to get his message across. He decided to write an email to Microsoft's CEO Satya to tell him that he was sorry, that he'd tried, but for whatever reason he couldn't get people to see the big picture. His lunch companions told him not to do it, but he kept writing.
“Then an email comes in, ‘ding,’” Marley said. “I look down and two threads below is Satya; he'd read a Reuters article and sent it to his direct reports, the entire leadership team.” The news agency had put out a story earlier that day with the headline “Microsoft Launches Cloud-Based Blockchain Platform with Brooklyn Start-Up.” Marley hadn't seen it.
“This is the perfect example of growth mind-set,” Satya had written to his deputies in the email. “And I was like, okay, we're here,” Marley said. “That's was how everything else all got started.”
Ethereum was now attracting significant interest from some of the largest and most powerful corporations in the computing industry. Banks were sniffing around the edges, growing more confident that Ethereum was a different beast, not tainted with the aura of scandal and money laundering like Bitcoin. What the project needed now was to prove what it could do. It needed applications, and it needed to keep growing the number of people who fell in love with its promise of a whole new world.