The secrets behind the success of Ethereum & Consensysby@ricruben

The secrets behind the success of Ethereum & Consensys

by Ric RubenJuly 29th, 2018
Read on Terminal Reader
Read this story w/o Javascript
tldt arrow

Too Long; Didn't Read

With Co-Founder, Joseph Lubin!
featured image - The secrets behind the success of Ethereum & Consensys
Ric Ruben HackerNoon profile picture

With Co-Founder, Joseph Lubin!

We had the privilege of interviewing Joseph Lubin, co-founder of Ethereum, founder of Consensys, as part of the Consensys Tel Aviv community and Investor events.

You run quite an amazing organization and have a forward-thinking approach on how to manage it in a decentralized fashion and at scale. It’s often referred to as The Consensys Mesh, can you tell us all about that?

We consider ourselves an unusual kind of company. We formed about three and a half years ago, about a year into the Ethereum project. We started as a venture production studio, and the idea was to build MVPs, wrap companies around them, put them up for external investment and grow. We got interest early on from companies, and governments, and central banks, and we needed to educate people in our ecosystem. We ended up doing different rounds of capital markets activities, so we basically have evolved into a very oddly shaped organization. We do a lot of things together that you don’t find as a single organization. We also started the thesis that the nature of business is evolving, so, it doesn’t make sense to try to build a big hierarchical, top-down command and control monolith like Microsoft, just because information is transmitted so easily and rapidly, expertise is shared, entrepreneurs are prevalent and they’re formed inside of these companies. Look at Google, they’re just so many entrepreneurs that they even do their own thing, and they remain part of alumni.

So we felt that the nature of business is changing and that we should embrace that, and create a context where entrepreneurs and serial entrepreneurs are attracted to Consensys, and there’s no better place in the world for them to do what they want to do. So we have evolved as this mesh of different product projects, service circles, and regional offices. And we’re attempting to be as decentralized as it makes sense, and that’s very decentralized using our technologies. The fact is that we’re currently in 13 countries while doing our best to be as flat and fluid as possible as an organization. Rigid hierarchy parities are great for different things, and they’re efficient for many things but not so good for decision-making. So you normally have the rich data and the situated awareness at the bottom of a hierarchy, and decision-making as far away as possible from the people who are making the decisions. We do our best to make the decision-makers the people who are in the different projects and give them budgets, so they’re really extremely autonomous. Yet we’re building an ecosystem of Consensys products and services we’re building on the Ethereum ecosystem, and we do a lot of infrastructure work, developer tools etc. We’re building out the decentralization ecosystem, essentially web 3.0. When you do that you need a lot of communication, collaboration inter-operation between different projects that we work with hard on, even while we’re trying to create this mesh of autonomous projects and companies.

We didn’t really set out to build an oddly-shaped company, it definitely emerged organically, and I think there are millions of people on this planet at this point that understand the value propositions of the blockchain technology. A lot of them are attracted to organizations, like Consensys, attracted to working in the blockchain space, at a company where everything is basically an experiment, they’re willing to iterate, and evolve pretty rapidly.

I’m wondering in this type of decentralized organization, how do you personally stay in the know and prioritize, because you have so many projects going on and so many different divisions, how do you personally stay on top of things?

The company’s growth has been exponential, the incoming interest is exponential, the opportunities are growing exponentially. It’s not just me experiencing that, it’s virtually everybody in the company. So we all have to re-prioritize what we can pay attention. We are all building, system processes, larger teams, in order for us to handle decision-making and information gathering better. We do meet one another, we have about 850 people around the world, and we meet one another very regularly at different events. There are lots of opportunities for us to spend good quality time together, work on stuff together, bond, and establish a shared understanding language so that when we use our communication tools, like email and Slack, we hopefully understand each other a little bit better.

My dad always says that in a company’s culture, it’s better to ask for forgiveness than to ask for permission. Is this your approach as well, at Consensys, since you’re talking about this experimental environment where people are exploring new ideas?

At Consensys there are lots of opportunities to ask for forgiveness. There are a lot of outspoken, passionate people and we practice Stigmergy. This is the phenomenon where if collective intelligence identifies some sort of goal collectively it is included into the species DNA in a sense, just like ants move sand around, and push little pieces of sugar around for the collective good. At the organization, we see pieces of work, and we push them forward in different ways, and people are encouraged just to act if they see something that should be done, and they do it. However, if the thing they want to do could impact other people or other potential goals, they should try to bring those potentially affected into some sort of discussion, get help or just let somebody know that something’s going on.

The company is called Consensys, and so we do try to form a consensus amongst anybody who would be impacted by some event or some decision, and some of those decisions impact the entire organization. As I said before, there are lots of opportunities to ask for forgiveness. We’re trying to see ourselves as an organism, and we’re trying to build immune systems into that organism, so we have a group that is helping us with communication protocols, such as non-violent communication techniques, emotional intelligence, and different kinds of programs to improve us all as operators within the Consensys context.

You also have something called the Traditional Management Nullification Tools, could you please describe what this means?

TMNT is a really cool project. The organizational philosopher and theoretician, Ken Wilber, defines different ways that humans can organize themselves for collective action. He labeled them with colors from Red to Teal. So a Wolfpack has the red designation, and army and machine, and family and organism are the next level of evolution in how we get better and better as a collective intelligence. At Consensys, we’re trying to pattern ourselves as an organism, so all the people constitute the elements of organs. We have product teams, service circles, and regional offices that are the various different organs. We’ve got feedback loops going on inside them, and we’re trying to figure out how we measure, and how we apply mission and goal setting. It is essentially intended to be healthy at the cellular level, where all the people are, healthy at the organ level, and ideally emergently healthy at the organism.

You can think of Consensys as one of these organisms, we have spun out a bunch of companies, they’re still part of the Consensys mesh, some of those companies are becoming hubs in their own right. We have what we call a hub-and-spoke system, where our different product teams are spokes to the Consensys hub. We have a team, called Labs that is working with all the different product teams to wrap them in APIs, and then assigning service level agreements for those APIs, so that you know what a team offers, and you know under which terms you can get those services. We also see our marketing team, our HR team, and our finance team, and maybe our legal team, as potentially proto-spokes, they may evolve it to a business in their own right, as we write software to enable them to do their job in this new decentralized context.

We have this technology called Open Law that we’re just starting to apply inside our organization. It enables you to write employment contracts, NDAs, and service level agreements with these hybrid blockchain-based legally enforceable documents. These mechanisms enable agreements to be formed and signed, you can sequence different actions, you can send money into a contract, escrow money in the contract, pay money out of the agreement, and send data into a service level agreement, get paid out based on the data and based on the performance of that particular team. Using these mechanisms, understanding that different companies that we’ve started and spun out are creating hubs on their own you can start to see this mesh.

Consensys runs over 40 projects, even more, at any given moment. Are there pillars or certain topics that you can categorize and say that these are the things that you believe are most impactful at this stage for blockchain technology?

It’s still an immature space, and we have lots of projects that are maturing rapidly, there are thousands of projects out in the world that are maturing rapidly, so they are more building blocks. There’s self-sovereign identity, that’s going to be extremely important. It’s going to be the way we access services on protocol-based open platforms on the decentralized World‑wide‑web: Web 3.0. Reputation systems, bounty systems, governance tools, accounting systems are all important components. They’re really useful in their own right, but they’re going to form components of more complex solutions, and other elements include standards for fungible tokens and non-fungible tokens, and atomic swap mechanisms, decentralized exchanges, and getting rid of counterparty risk.

Applying those same ideas to the insurance industry or to lending, where instead of standing up an insurance company that offers products, we, and others, are trying to build open networks that are protocol-based, so that anybody can put a parametrized insurance product on this protocol, anybody can put a parametrized lending product on this protocol. And trying to build as many of these standards on Ethereum and push them down in the stack to create this shared infrastructure that we can all collaborate on and enable people to essentially get a lot of that for free and build useful products for different consumers and businesses.

Essentially, we’re moving from a world where we had monarchs or sovereigns setting the rules, making the judgments, subjective governance, to a world where we increasingly have rule-based governance. However, that rule-based governance, in countries where that applies, is usually subjectively determined, defined and applied and executed. If we can take that rule-based way of organizing our society and we can move it from subjective manual application of trust to an automated guaranteed application of trust of these rule-based systems, then we will have a better foundation on which to layer all the technology that we have built over thousands of years, and hopefully a more equitable, more decentralized government, and a better society.

Let’s talk about Token Generation Events. In 2017, utility token was the buzzword, and now in 2018 we’re all sitting and talking about security tokens. I’m wondering what your take is? Because in every project you guys also have a token, you have Token Foundry, you’re building new economic models, so how do you see the evolution of all this?

The Token Foundry team has helped some of our projects, and third-party external projects to token launch themselves. We’ve issued investor tokens, one great example is a project called Brade. It’s a movie that got into Tribeca, and it shows in the Tribeca Film Festival, so it’s a very exciting project, and we essentially crowdfunded a movie. The people got tokens for that, and we raised 1.7 million dollars, it’s a security, and it enables people to get 15 percent back on top of their principal, they’re getting 30% of profits in perpetuity, so it’s a cool system and an experiment.

We have many different projects that are consumer utility tokens and would not be considered securities, even in this context, at this point in time, there’s a lot of fear, uncertainty, and doubt that has been raised by regulators around the world. The SEC has made certain pronouncements about consumer tokens or utility tokens, early statements were to the effect that all these tokens are probably securities.

Regulators around the world have a dilemma, the problem is that there are a lot of great projects there, a lot of fraudulent projects, a lot of projects that are selling what could and should be good, well-constructed consumer utility tokens, but they’re selling them to investors in enormous quantities, in quantities far greater than they could actually use, so they’re really selling them as an investment. So the SEC’s opinion, from what we’ve discerned, is that those things constitute securities.

So we have a framework that enables us to sell consumer utility tokens, not in enormous quantities, not with discounts for large investors, not to investors you essentially have to be a technically accredited token buyer, you have to answer questions, you have to demonstrate you will use these tokens on these platforms, and so that that’s a good clean way to sell a token that represents some sort of access or consumption of a resource.

There have been the nefarious actors for thousands of years who take advantage of information asymmetries in order to defraud consumers, and that’s still going on in our space. I think that we will have the tools to make it a better situation. Unfortunately, the global context of this technology and the low barrier to entry for running a token launch has exacerbated that information asymmetry. But we can make use of those characteristics to self-regulate and to build systems that protect consumers better. In the old context, you can register shares and sell them using a bucket shop to naive grandmothers via the telephone. That’s a hard scheme to ferret out in this world.

In order to have a successful token launch, you have to publish a white paper, you have to get a telegram or get a website up, and we can find those things pretty easily no matter where they are in the world. We are creating a project called the True Set, that’s an EDGAR-like database for token launches. Ideally, if you’re going to be a legitimate token launch, you’re going to have to register significant data in this database, so we will all be able to point to it in the database. If possible, we can have people scrutinize that database and determine that the disclosures are accurate, and so we’ll have this wisdom of the crowd dedicated to identifying these projects and analyzing them.

We have a project called Frontier that incentivizes professional analysts to share their work. If you look at platforms, like Reddit, where people just tear apart white papers, there is a lot of energy out there that can be directed to cleaning up this space. We certainly need bodies like the SEC to scare many projects straight, and to create a context of concern so that many different projects think twice and do their legal homework before releasing their token.

In Israel, we’re considered the Startup Nation in the traditional startup scene, and we’ve recently been acknowledged as the Crypto or the Blockchain Nation. You’re about to open an office in Tel Aviv, I’m wondering what is your vision or your strategy and what are you on the look when you open a new branch somewhere in the world?

Over the last three and a half years, we’ve gone places that wanted us, we spoke to people who wanted to learn about what we’re doing. We were happy to speak to anybody who wanted to listen to this stuff, now we’re pretty overwhelmed with demand, and we’ve now done various different projects around the world in different nations, with different central banks and governments.

But two places in the world feel a little bit similar to me — one is Israel, because of the startup culture, because of the world-class cybersecurity, cryptography, and mathematics expertise. The other place is Silicon Valley, where you have technologists who are building the systems that they believe effectively structure the world. Those two places, especially Silicon Valley, was slow to embrace Blockchain because of course, they know how to run the world based on Facebook and Google and other technologies. They are a pinnacle of the web 2.0 world. But some early adopters moved into the Bitcoin space, and they iterated in that space for a while, and I think they have now realized some pretty narrow technology. It’s really just one application from an infinite number of applications that you can build on the blockchain. So, in both those regions we felt that the people there have extreme talent but also were very sophisticated, and we should up our game, both at the Ethereum platform level and at the Consensys level, so that we could go into those places and answer all the questions and have respectable things to say in those particular situations.

So a year ago, we formed an office in San Francisco, it’s grown to about 50 people now. We’re about to take a much bigger office and to really be a catalyst. Ethereum is starting to catch fire there. It’s being aided by GDPR, it’s being aided by Cambridge Analytica, and lots of engineers are just wondering what they’re doing with their lives. That business model of exploiting people’s personal information may not last that long, and it should evolve.

Israel is really exciting to us because the talent here is tremendous. We want to build an R&D lab here like many companies do, but we really want to build a full-stack hub. Some of the earliest technologies in the Bitcoin space, in the early colored coins, and Mastercoin projects were performed here. Ethereum is gaining quite a bit of traction here, as people realize that it is a pretty powerful technology, a decentralized application platform. Full stack to us means we build an office that focuses on education, through Consensys Academy, we hire lots of engineers, they can do consulting work, we bring projects into the Mesh and incubate them, we invest in projects in the ecosystem. All elements of human society are invited to come in join us in the decentralized future.

Originally published at on June 8, 2018.