Behind the scenes with Tezos, a new blockchain upstart

Odds are you have at least one buddy who bought bitcoin a year or two back. Odds are that they are giddy about their good fortunes. And they should be–since Satoshi launched the network in 2009 bitcoin has risen to a market cap of $40 billion. The cryptocurrency saw a 200% surge this year to a record high just shy of $3000 in June.

But beneath these numbers, the feuds among core developers over resolving longstanding issues have simmered on, most critically how to increase block size. Under the current 1MB block capacity, congestion of the network has put bitcoin under strain, resulting in higher transaction times and processing fees. Two of the top scaling proposals are set to update the network on August 1st, dubbed Bitcoin Independence Day.

The community is bracing itself for the possibility of a chain split if the proposals end up being incompatible, an event that would result in two separate bitcoins. The players in this drama all have incentives to reach a consensus, which begs the question–how did we arrive at this juncture? One answer is that bitcoin’s key feature and strength–decentralization– also makes decision-making remarkably difficult.

With no central authority or mechanism to get participants with different interests and ideologies to agree on changes, bitcoin and new platforms like Ethereum are struggling to amend their core rules. Math may be math, but humans are also human.

Bitcoin will no doubt get through these trials and and hit a market cap in the trillions. But the civil wars raise important questions. Enter Tezos. The upstart platform is designed with an eye towards avoiding these flaws by baking governance into the protocol itself, allowing for formal coordination and upgrades.

In an interview for Flux I sat down with Kathleen Breitman, co-founder and CEO of Tezos, which is nearing the end of it’s two week fundraise.

With a whopping $200m raised (at current btc and eth prices) Tezos breaks all Initial Coin Offering records to date. That’s saying something given over $1 billion dollars has gone into ICOs (Kickstarters on steroids) in 2017. Exuberance, of the irrational variety, perhaps–but that doesn’t mean there is no real value underpinning some of the projects raising funds, or in the underlying crypto-token model.

RRE is not involved with the company and this interview is not a recommendation to participate in their fundraiser, or in any ICO, but hopefully it will take you beyond the headlines and into the live thinking of a startup on the forefront of this trend.  In the interview Kathleen discusses the rationale around doing an uncapped fundraise, how the team is navigating a landscape with no existing regulatory framework, why they chose to set up a non-profit in Switzerland to receive the funds, and how they have responded to DDOS and phishing attacks. Kathleen also discussed her identification as an anarchist-capitalist and its roots in cypherpunk, her love of Milton Friedman, and the quirks of being a husband-wife founding team.

An excerpt of our conversation is published below.

AMLG: I guess we should start at the beginning and explain the origins of Tezos. You’re the CEO and co-founder along with your husband Arthur. Husband-wife team and he’s the CTO, the architect. You told me you guys met at a crypto-anarchist meetup in New York in 2010. What is the story there and what kind of pickup lines does one hear at an anarchist meetup?

KB: It all starts off in a in a bar near NYU in 2010. I was de-briefing with my debate team. We had just lost mightily to a better school in a debate. There was a party going on at the bar and I was 20 years old and was very tempted at this party by the free-flowing beer, a hot commodity for 20 year olds. I met this guy named Tennyson. Tennyson and I got along just fine and he insisted I come with him to a lunch the next day. He left out a lot of details. The deal-breaker probably would been that it was crypto anarchist lunch — minor detail. The theme of the lunch was not disclosed.

Arthur and Kathleen Breitman

So I went in and this guy made a bee-line at me. I was a little weirded out but Tennyson had told Arthur all about me prior to this and I guess he thought that it would be a good match. So shout out to Tennyson. He said “hello, nice to meet you, welcome to the media.” He was friendly and gregarious and tried to figure out all the things I would like and not like. He name-dropped that he knew Milton Friedman’s grandson when I mentioned I liked Milton Friedman. We just kept on talking and I really enjoyed speaking to him and he seemed smart and nice and funny. I meandered through the rest of my Saturday the way I would normally. Then he texted me later and I texted him and we wound up meeting at another bar. Then we wound up going to dinner —

AMLG: And you covered all of Milton Friedman?

KB: Oh yes it was a rich intellectual history of the Friedmans. Basically we just started talking and we haven’t stopped.

AMLG: And Arthur was a quant right. He was at some high frequency trading shops, shops being Goldman.

KB: Yes. He studied math and computer science and physics over in France. Then he wanted to work in finance. France isn’t a great place to make a living if you’re under the age of like 50. So instead of going to London which he didn’t really care for he decided to make his bets on New York. So he went to NYU where he studied under Nassem Taleb and the financial engineering group. Then he became a quant. So he was doing high frequency trading and market making for foreign ETFs in particular.

AMLG: You’ve said that Arthur is the brains and you’re at the brawn. What do you mean by that?

KB: Arthur is really the technical architect, he came up with the structure. Tezos has gone in and out of fashion — the ideas — over the last three or four years. People really took interest in it last year when the DAO occurred. This was a fantastic heist of sorts. A digital heist.

ALMG: Could you explain what happened there, with the DAO hack?

14% of all Ether was deposited into the DAO, a smart contract (computer program) running on Ethereum. It became at risk due to a flaw in the contract

KB: There was a project that aspired to be a decentralized venture capital organization on the Ethereum blockchain called the DAO. They had about $150 million dollars worth of Ether pledged to the contract. This was an ambitious undertaking and the people who created the DAO didn’t do a super thorough job of security audits and wondering if they could actually mobilize this vision they had created. So a clever hacker gave himself some of the Ether out of this contract to the tune of $50 million dollars out of the $150 million.

AMLG: That’s a good steal.

KB: Not bad if you can get it. So this forced a conversation in the Ethereum community as to whether or not there ought to be a do over. Call it a mulligan, if you will, on this whole DAO fiasco. That led to a split in the chain of Ethereum, which is how we got two versions: Ethereum Classic which is favored by some people in the community because it didn’t unwind this transaction — it forced people to deal with the consequences of this heist. And then regular Ethereum or Ethereum as we know it today, which basically called a do over. This was ultimately a governance question and the whole nexus of Tezos is — governance is very much a part of blockchains, we can’t divorce the math and computer science aspect from the overall governing philosophy. So this forced a conversation.

AMLG: So when there’s a fork and a community has to vote on a direction there’s basically an informal process that’s determined by perceptions, perception of which fork seems most authoritative or valid or which constituents shout the loudest which is what we saw after the DAO hack. This is the problem you’re trying to solve with Tezos — you’re trying to formalize the process of how one upgrades a protocol?

KB: So the Tezos position paper examines the potential threats to bitcoin. Ethereum was not factored into the analysis in the initial white paper or position paper, but yes essentially the idea is that ultimately these systems are very fragile in the sense that if you have undue influence over the protocol on this soft level you can render different results at the protocol level.

You see this with bitcoin. You have core developers who have personalities and it kind of becomes a cult of personality over who you’re with and people taking sides on Twitter. But there’s no meaningful way to enact change. In the sense that ultimately you’re at the behest of miners in China over which version of the protocol is actually deployed.

AMLG: So your mechanism allows blockchain to change the blockchain itself. Very meta, turtles all the way down there. Your central idea was — you saw this fallacy underlying bitcoin, that code can solve everything. Or as Arthur puts it, nothing bad can happen to bitcoin because MATH. But really blockchains are embedded in society, society can be compromised. So this is what you’re getting at with Tezos right. The question I have that follows — under the current construction isn’t there a risk that the network favors the people who are more technical, the more technical stakeholders who are knowledgeable about anything that you would vote to change? What do you do for the non technically-savvy tokenholders?

KB: That’s a good question. We have what’s called delegated proof-of-stake in our system. Essentially what this means is that everyone who owns a token can become a validator in the network. Or they can actually assign that right to someone else and that person can vote on their behalf. So there is going to be some element of politicking, but at least it’ll be something where people can express that in a much more legitimate way than just saying that that guy’s right because —

AMLG: So sort of a proxy vote.

KB: It’s a proxy vote yes. Some people call it liquid democracy.

AMLG: But you can never really remove the element of humans influencing other humans’ ideas right, it’s just the voting piece?

KB: Most definitely. That’s an important caveat. That’s why it’s an experimental piece of technology. Ultimately this is still going by people’s judgment, which is of course always subject to some sort of fallacy. The main innovation with Tezos is giving a way to air legitimacy on the chain. Right now it’s sinuous as to how decisions are made. By making things more explicit we hope that we can get more useful conversations going.

AMLG: It’s super relevant now since we’re seeing this play out with massive battles over block size in bitcoin, whether to go with bip-148 or SegWit2, all these different things. Maybe you could just fit Tezos into the landscape for the listeners. You’ve said previously that the Bitcoin model is so decentralized that it’s beholden to these populist campaigns and there’s this governance deadlock. Then on the other side, Ethereum is moving very fast but is very centralized because the Foundation has a lot of clout over the protocol. Where do you fit on that spectrum?

KB: In some ways we’re choosing a middle ground between the two and in other ways we’re taking a radically different approach. So a little about how Tezos governance works in its first iteration: what we’re going to do is a two-phased vote. We’re taking our cues from the Republic of Venice and we’re doing affirmative voting on the first iteration. So what would you like to test. So say six people have proposed these things and they all have different tradeoffs, what we want to actually test and see if it’s useful for the protocol. What’s the proposition here.

AMLG: Then that proposition has to pass a threshold to get the main vote?

KB: Yes after testing. We’ll spin up a testnet. It will say for example, well you said that this preserves privacy but we actually found that it does not.

AMLG: What happens if you don’t get people over a certain level? Some of the DAO propositions won only 5 or 6% the votes.

KB: One way of approaching that is delegation — making it easy to not have to think critically about every proposal. If you take a vacation you shouldn’t have to worry too much about the network.

AMLG: Right and it’s very easy to bash every option down.

KB: But we also adjust the quorum every time. For instance maybe people just want two people making all the decisions. That’s a weird outcome certainly but it’s possible.

AMLG: Is it possible that Ethereum could change its governance structure?

KB: They could introduce different roles for people. They could introduce something like a representative democracy. But the innovation with Tezos is that we can actually iterate on our governance model. We can adjust it at the meta level. That’s something you have to bake into the protocol at the base layer.

AMLG: They really can’t bake it in at this stage — it’s done?

KB: I think it would be remarkably difficult. I’m not going to underestimate Vitalik of course. But with Tezos we don’t just have this governance layer, we also have a way of pushing the upgrades out to everyone, it’s one and the same, making it a very powerful mechanism. Whereas bitcoin and Ethereum still have manual processes for updating their chain.

AMLG: You’ve said yours is modular, with the three layers swappable you can swap things in and out?

KB: Yes. It’s called a hot swap on the protocol.

AMLG: Hot swap on the protocol. Well Vitalik is a brilliant guy, but I did love the fact that your smart contract language is Michelson— 

KB: Michelson. At least that’s how French developers pronounce it. I actually don’t know, I mispronounce everything —

AMLG: How’s your blockchain French?

KB: Better than my blockchain English.

AMLG: But so it’s a reference to the Michelson-Morley experiment, which disproved Aether right? Not to be confused with Ether. That’s hilarious. I love that you guys throw those little things in.

KB: It’s the only cheeky thing we’ve done.

AMLG: The other interesting thing is that you’ve built in incentives for developers — bounties and invoice potential on the chain.

KB: Yes. So I think a big failing of open source technology nowadays is that you basically have corporates sponsoring it because they use it in some capacity and that’s kind of how these things are governed. But outside of a group of enthusiasts and people who have a vested interest in its success you don’t find too many people actively trying to contribute just for the sake of it or for pecuniary gain. We decided early on that that would be a massive inhibitor to innovation in these blockchain protocols. Because at the beginning at least they seem to appeal to ideologues — which mind you is just fine — but I like the idea of having someone from a university who’s been studying something on the periphery and incentivizing them to contribute to the protocol, even if they don’t find the ideological aspects of it appetizing.

AMLG: For the mercenary engineers out there.

KB: There are many of them.

AMLG: Really you guys have been working on this for a while but it feels like you’re just making your entrance onto the scene more. In the grand scheme of course it’s still very early, but surely many of the folks interested in the space have already taken stakes in other competing ledgers? How have you navigated these existing vested interests and converted people to the Tezzie religion? I know you’ve said that community building has turned out to be one of the hardest parts.

KB: It’s genuinely really tough. I think that’s part of the reason we did an uncapped crowd sale. When we were thinking of the best way to get Tezos backing from people, it wasn’t obvious to us how to do it. But one way is to make sure that everyone has the ability to take part in the ecosystem. People having an interest in you and contributing to your fundraiser is a way to get them interested in the technology.

AMLG: You really opened it up broadly with the fundraiser.

KB: Absolutely.

AMLG: Should we not be saying ICO?

KB: I mean it is a fundraiser. People call it an ICO. But it is a fundraiser. I was thinking of making tote bags like PBS tote bags.

AMLG: Because we’re donating to a nonprofit.

KB: Exactly. And the non-profit is recommending an allocation of tokens.

Roughly ~$1.3b billion cumulative dollars have gone into ICOs in 2017, with over $600 million in June alone. [For more data see Smith + Crown.]

AMLG: Let’s step back for a second and lay out the landscape for people who haven’t seen all the headlines on ICOs. What’s a quick explainer on an initial coin offering and what have you done so far.

KB: Basically there are a number of new projects in this space and one way to distribute the token is to simply receive donations to their blockchain and then in return you get the token. And typically they’ll price it at a certain number of Bitcoin or ethereum and deploy it right away. So it’s an alternative financing structure for some projects and for some projects it’s a distribution mechanism

AMLG: And the tokens are not securities.

KB: I mean in some cases they could be right, like if you have a claim on a line of revenue. Arguably that’s a security. But with Tezos you’re basically getting computing power on a decentralized blockchain.

AMLG: Let’s get into some of the questions that have been raised — we ought to cover a few. I admit when I first heard some of the facts about Tezos I was skeptical. It’s easy to fall into cynicism. There’s all these keyboard warriors on Reddit throwing out one liners. But the papers you’ve put out actually give clear and well-thought out answers to some of these mainstream public questions. I want to get into a few of them. The first thing I want to ask is about the pushback over the fact that the fundraiser is uncapped. Uncapped meaning there’s no upper limit on the funds you can raise. The limit is just the time, the window in which you’re raising — two weeks. In the Tezos overview you state the reason for this which I will read here:

“There is no cap on the amount of contributions that will be accepted. This is done in order to ensure that participation is not limited only to insiders or the fast fingered. The Tezos team believes that an uncapped fundraiser will promote a widespread distribution of the tokens a necessary prerequisite to launching a robust network.”

So that inclusive argument instinctively makes sense to me. But of course the problem with democracy is democracy. The system itself. Crowds don’t always get things right and in this case your crowd may be speculators. There’s been a lot of speculative buying in China for instance. Do you worry about who the tokenholders are and whether their incentives are aligned with Tezos?

KB: It’s tough. We didn’t do much marketing outside of the U.S. Well rather Arthur and I are based in the U.S. and we talk about the technology in mainstream U.S. press outlets and sometimes it’s picked up in Asian outlets. We mostly have just been evangelizing about the technology. But certainly there are a lot of people who are interested in the more speculative aspects. I think there’s a lot of fervor and froth in the marketplace right now. That does make it a bit odd to launch something that’s more community-based on some level when there’s a lot of people who are just profit-seeking. I suppose I do worry but ultimately we’re appealing to people’s rational self-interest. So at the end of the day I hope that the proper incentives will align themselves.

AMLG: And that the tokens end up in the right hands, with people who align with your values?

KB: Yes.

AMLG: Let’s talk about the structure, which is the other question people ask, and fiduciary liabilities. You’ve laid out the Tezos Foundation’s goal is to promote the use of the Tezos protocol. So participants in the crowd sale donate to the Swiss Foundation which in turn provides the contributors with a token allocation. You’ve said the Foundation will spend on conferences, marketing, funding developers. If your director decided he wanted to buy an island in the Caribbean who votes on that? How does the corporate structure work?

KB: That’s a good question. Part of the reason this is not a security is because you have no claim on the revenue and that includes the revenue from the fundraiser. The Tezos Foundation is at the behest of the Swiss nonprofit authority, which has a clear directive to adhere to whatever the Tezos foundation sets out to do. So if Johann wants to buy a Caribbean island where there there’s a ton of Ocaml programmers that’s kosher if Johann wants —

AMLG: So if they program better there it’s OK — as long as productivity goes up.

KB: Ha yes. If Johann wants to buy a Caribbean island because he wants to run it on the Tezos blockchain in some way, I guess that falls in line with it. But if he wants to do it because — well he’s quite pale so I don’t know if he wants to go tanning — but if he wants to go on vacation and buy an island that would not be OK. You could theoretically hold him accountable under Swiss law. And I think that the Foundation ought to survey the community for major decisions to reflect their will better. But there’s no explicit voting mechanism on it. Part of the reason that’s not such a bad thing is that anyone can contribute to the protocol. There’s the Foundation which has veto power for one year, but other than that they don’t have any favored status in the protocol.

AMLG: It’s amazing how you’re paving the way and there doesn’t seem to be any regulatory framework for what you’re doing. What was your thinking on going to Switzerland? I know that the U.S. money transmission laws are nuts on a state by state basis, it’s so complex here. But it sounds like it’s complex there as well. What are your thoughts on how the regulatory framework will evolve?

KB: Well Johann is the founder of the Crypto Valley in Zug, which is a Canton in Zurich. It has a ton of Bitcoin companies it has a ton of watching companies located there and including Ethereum foundation.

AMLG: So I guess they paved the way. Do you think others will follow — do you think there’s a best practice being set here?

KB: I think so. There are other a few other places — Gnosis which is an ethereum prediction market token I think they had their token sale out of Gibraltar. Estonia wants to do work around this. I think the Ethereum foundation itself is actually in Singapore right now. So there’s all sorts of accommodating usually small nations that are willing to work with you.

 

AMLG: Why is it there so much nastiness in the blockchain crypto world — is it just smart people with opinions strongly held? Or is it monetary reasons?

KB: I think people are just naturally inclined towards taking sides, you see this in politics too. “I’m team Ethereum. Ethereum is going up 4,000X why don’t you like ethereum, why do you think you’re better than Ethereum?” Right.

AMLG: But actually it seems like you cooperate with a lot of the other folks in the space. I mean Zcash, Zooko — he’s one of your advisors right?

KB: Oh yes Zooko. In general we have a good working relationship with most of the people in the space that we admire. It’s really folks who can’t evaluate our technology who get really upset with us and start to caterwaul. You can’t let that get to you. Honestly if you look back at Ethereum people were much more vicious. So we got off pretty easy on the vilification front. Most of the attacks have been very personal. But I’ve been on the internet since I was 12. And I worked at Bridgewater so I have thick skin.

AMLG: All the character building you needed.

KB: Exactly. You’ll have to come at me a little harder if you want to hurt my feelings.

AMLG: How do you cope with the stress, the two of you. I guess you’re both moving around internationally and busy, but you are in a relationship with each other — 

KB: I mean, it’s compartmentalizing. Sometimes it’s “OK right now you’re my business partner and I’m telling you that this is a terrible idea.” And sometimes it’s “I know you’re stressed out. We should really talk about this. But let’s first see a movie or watch bad TV.” I think we just know each other really well.

AMLG: What is your go-to bad TV?

KB: The new like 24 spin off, Designated Survivor. It’s just so awful.

AMLG: Your brain doesn’t even need to be switched on.

KB: It’s just its so stupid. It’s really fantastic because it’s like conspiracy theory after conspiracy theory and you can’t quite trace it.

AMLG: And your parents now know that you’re doing this right? I know you were “a financial software engineer” for a while.

KB: Yeah I was doing “accounting software.” Now they’re on to me. My mother unfortunately has learned what Reddit is like. But she still loves me which is great. She doesn’t seem to be assuaged by the arguments that I’m an incompetent boob or just terrible. So I’ve got that going. My mother still loves me.

AMLG: I’m curious to zoom out for a minute on your self-identification as an An-Cap. What exactly is an anarchist capitalist, in rough terms?

KB: In general its someone who believes that markets do a better job of distributing goods than the government. They tend to have a preference against government aggression in any form.

AMLG: You put at the top of your white paper “laissez faire les proprietaires” which sums up a lot of this. You’ve said the whole cypherpunk movement is about using the Internet to free people both informationally but also financially. It’s interesting to think about how Tezos and the work you’ve been doing builds upon decades of thinking. Privacy advocates have been organizing in Silicon Valley since the 80s — guys like David Chaum, Timothy May and that cypherpunk mailing list. Which then evolved into the cryptography mailing list where you got started. I have a quote from May’s manifesto which I thought would be fun to read: 

“Just as the technology of printing altered and reduce the power of medieval guilds and the social power structure. So too will cryptologic methods fundamentally alter the nature of corporations and of government interference in economic transactions.”

That was 1988 — 20 or so odd years before Bitcoin. People have wanted this for a long time and then cryptocurrencies raised their head again in the mid 2000s and then boom Satoshi. Where do you see yourself fitting into this long history of crypto anarchist capitalists?

KB: To your point it’s been a series of failures and a series of people who are a bit too far ahead of their time. So I hope my timing is right. It is just massively experimental. So we have to take that into account. Bitcoin is the new fad. It’s the new fashion. People like to talk about it because it invites people to think about a world without intermediaries as they know it. That’s cool. Maybe bitcoin isn’t the end game. Maybe it’s some other application that’s built. But I think it’s a very inspirational piece of technology.

AMLG: Where did it start for you? Was there an Atlas Shrugged moment when you were young. Did you read some book?

KB: My parents are both pretty far to the right and classical conservatives. My dad subscribed to National Review when I was little and it was like read this. So I grew up with a more right-leaning bias. Then I read Milton Friedman when I was like 16. He died in 2006 when I was 16 and I read his obituary written by Amity Shlaes. It was eloquent and talked about how Friedman really cared about people’s ability to prosper and make commerce for themselves and that was a nice idea. I think instinctually I don’t like being told what to do. I don’t like any form of tyranny. Even the more mundane types like being told to go to class.

AMLG: I was on Twitter this morning and I thought you would enjoy this article that Andy Weissman of USV tweeted out. So a prominent theoretical physicist, Yaneer Bar-Yam, who crunches big data and generally analyzes complex systems — he actually accurately predicted the Arab Spring — he borrowed mathematical frameworks from quantum field theory and did a study on government. He found something that we have already been thinking about — that hierarchal systems no longer work given how complex society has become and how globally interconnected we are. Hierarchies used to work in smaller kingdoms but since the 1980s there’s been a breakdown. Whether it’s a financial crisis or environmental issues or diseases, it’s just too complex for a few entities or people at the top to respond to and decide on policy. At the end of this analysis he proposed a shift to “horizontal distributed governance” which to me points straight to you.

KB: Why didn’t he publish this like six weeks ago.

AMLG: He also looks at the brain. If you think about the brain there isn’t one neuron telling the other neurons what to do. There’s a collective behavior that’s responsible for how we think and decide. As you read this research you think OK this makes sense, everything is too complex, but then what’s the mechanism?

KB: Yeah part of the reason that we set up Tezos the way we did — and we could have been way more prescriptive with the initial governance model — the reason we chose a vanilla two-phase vote was because we didn’t want to be too prescriptive. It took society thousands of years to come upon the institutions that we have today. There are subtleties that you just can’t quite capture. You can’t capture them for a variety of reasons. Part of the limitation is that Tezos is going to be digital commonwealth. That means it’s not going to be particularly tangible outside of the world of the Internet. But also there’s a lot of counterintuitive things when you look at government and the way it works today and the way liabilities work.

One of my favorite books which has just been updated is David Friedman’s “Legal Systems Very Different from Ours.” It talks about things like punishment structures in different societies and cultures. It’s interesting because you’re always trying to optimize for one thing and you’re not always super sure what types of incentives you’re building in. One of my favorite anecdotes is that in the 1970s in Vietnam they used to give money out for rat pelts. The idea was they wanted to get rats out of the cities. What happened instead was that there were elaborate rat farms. You can’t always fine tune human behavior. You never quite know it’s going to happen. So it’s best to take a wait and see approach and see how communities emerge and how they self-organize.

AMLG: Why the name Tezos?

KB: When we were coming up with the name Arthur wanted an original .com so he wrote an algorithm to find an unclaimed .com that was pronounceable in English. Apparently it’s slang in Colombian teenager speak on Twitter for “expert.” So that’s a massive freebie.