Hashed Interview: Joe Lubin, ConsenSys, “What I See In The Market”
“Many companies and consortia are interested in decentralization for their internal systems, because they can work and collaborate with different companies in their space. Nobody has to own the machines or the software, and everybody can essentially share ownership and even intellectual property. It just makes it easier to build collaborative systems, whereas it has been much more difficult before blockchain.”
Kenny Kyungjin Kim, Head of Research at Hashed, interviewed the founder of ConsenSys Joe Lubin and covered several topics including DTPS, DAO, ConsenSys 2.0, and dApp.
Joseph Lubin, welcome to Hashed. Thank you for agreeing to an interview with us. It’s a pleasure to have you here. I’m going to ask you some questions regarding blockchain, cryptocurrency, and more. I want to start with your keynote speech today; you talked about a new concept called DTPS, which I thought was quite interesting. Can you tell us more about that?
I’m glad you’re interested — that’s great. So DTPS, decentralized transactions per second, is an idea that a bunch of us at ConsenSys have been talking about for a while. The idea there is that measuring transactions per second is both subjective and kind of meaningless in the blockchain ecosystem so we can have many transactions per second in a SQL database and that’s great, or we can have many transactions per second in a slightly decentralized blockchain system; that’s nice, but I think one of the main goals for the blockchain ecosystem is to get to trust, to create a revolution in trust, to build a much better foundational trust layer for all the things we’re going to build in society and so we should start thinking about how to measure decentralization and there are some efforts already — arewedecentralizedyet.com, etc, and once we have pretty good notions of the different dimensions of decentralization, we can start comparing what a transaction looks like in Ethereum versus what it looks like in a less decentralized system, perhaps, so first of all, transactions can be very different over different systems — you can have a very heavy transaction on one system, and a very light transaction on another, and everyone just measures transactions per second so it’s very difficult to match those things up and derive meaning from that. But most importantly, certain platforms are very decentralized, maximally decentralized — Ethereum and Bitcoin are very decentralized, and yet still they’re not as decentralized as they could be, and will be, hopefully, over time, so we should put a number on that, and a number on other platforms that claim they’re the most decentralized platform in the world, and compare them. And so the idea is that we’ll put up a website and we’ll invite the world to come and help us define the different dimensions of decentralization. And so in an ecosystem, you could have some dimensions — essentially what the foundation governance looks like, how many developers there are, how many clients there are — that may factor in, but it wouldn’t be as core as how many different nodes there are on the blockchain, what the consensus algorithm is, how many different owners or controllers there are for the nodes, and even subtle things like are all these nodes sitting in the same power grid? So are a million nodes in a single US-based power grid owned by a million people more decentralized than say five nodes that are in different power grids owned by five different people around the world? So we’re trying to source some information on that. The website will be released in a little while and source the different dimensions and what those dimensions mean and we’ll put together objective statements on how to put these measures together. So the idea is that you take well-defined transaction-per-second number and then a decentralized quotient and then you take the product of that. So if we have a standard transaction, we figure out how many of those run on Ethereum, and if we have a decentralized quotient, of, say, 0.7 out of 1, for the Ethereum network, that would be the DTPS, the quotient times the number of transactions per second.
So I think that TPS-wise, it’s clearer, but in terms of DQ, decentralization quotient, there could be some controversy.
It’ll be crazy — the DQ wars, everybody fighting over what the different dimensions are, the different definitions of the different dimensions; the whole point is to bring in lots of different people with different opinions and to start talking about this kind of stuff.
So the initial numbers you showed in the keynote —
We just made those up. I tried to say in the keynote that these are “hand wavey” arguments. These are not based on any real data or not even very much thinking. It was just one of the first presentations of the idea and just a sketch, really.
So earlier you said that Bitcoin and Ethereum have room for more decentralization. So what would that be?
So the biggest and most obvious one is when Ethereum moves from proof-of-work to proof-of-stake as a consensus algorithm. It will likely get much more decentralized, it will get more scalable, and it will get much more secure because it’s a better algorithm. Because people won’t have to buy expensive hardware, because people won’t have to pay a huge amount for electricity, and bc efficiencies of scale won’t apply in POS, so there won’t be any advantage to having a whole bunch of phones or little devices running POS nodes in a warehouse, so you running them on your laptop and me running on my phone across the world would be just as efficient. In that sort of situation, with the barriers to entry so low, anybody can run a validator on their laptop or maybe their iPad or their phone sometime soon so we think that’s going to bring many more people into the ecosystem. You’ll be able to stake 32 Ether if you want to do it directly or there are different pools where you’ll be able to stake less than 32 Ether and participate in validation of transactions and secure the Ethereum network. So when Casper POS comes online and it will — so Ethereum 2.0 is coming out in, say, 12–18 months, and the first phase of that, Phase 0, should be — there are 8 different teams — and they should be in sync on a test network within a small handful of months. Maybe four months. So very soon we’ll be able to run Casper POS. It won’t be a real blockchain system yet, but we can start to see how that system works and there will be real value tokens on that network. So I imagine when Ethereum 2.0 comes out, it will be by far the most decentralized platform on the planet.
I think the concept of DTPS is interesting, but at the same time, I’m thinking of other comparable smart contract platforms. So back in 2017, Ethereum was the only smart contract platform, right? Now there are many more. Do you think they will be useful?
I think it’s great. I think it’s validation. It means this is a good idea, and when human beings see a good idea, many of them rush in and they try to create variations of that idea or better ideas. Some of them just try to monetize different variations of that good idea. The technology — the blockchain technology, the decentralization protocol technology — is still immature and so we need many different projects around the world to explore the solution space. It’s great — there are some really great projects out there and very many not really great projects out there, and whatever happens it’s pretty clear to almost everybody that Ethereum will be if not the absolute dominant platform, one of the most important platforms for quite a long time. It’s by far the biggest ecosystem and it’s moving faster than any other ecosystem it’s bringing on a huge amount of scalability right now in terms of decentralized transactions per second so I think by the end of this year we’ll have hundreds of thousands or millions of decentralized transactions per second using layer 2 technologies or state channels or zk-SNARKs or zk-STARKs mechanisms, plasma chains, and etc. So Ethereum’s doing well, it will hopefully share some of its technology with other projects and also learn from some other projects. The Ethereum project has never been bashful about taking good ideas from elsewhere to incorporate it into our systems. And that’s the beauty of open-source and the blockchain space is competitive but it’s also pretty warm and collaborative. The Ethereum ecosystem is very warm and collaborative and we all see each other at conferences all the time, and except for a few projects, people are quite open about what they’re doing.
This might be a tricky question, but I’ll give it a shot. If you had to recommend a smart contract platform to the developers in the Hashed audience who don’t like Ethereum as of now because of the transactions per second or confirmation time, then which one would you recommend?
I would recommend a Layer-2 technology that’s linked into the base trust layer of Ethereum. So if you’re building a game, for instance, you can use the Loom technology — a game company called Axie Infinity is bringing their own scalability technology. So why choose an inferior system when you can get all the scalability you want in a less decentralized blockchain system, but when you link it into Ethereum with Plasma, you get pretty much the full security of the Ethereum network. If anything goes wrong on that game platform, if the game developers or publishers are trying to steal your digitally scarce sword and sell it or use it in some other game, they can’t do that. Essentially, Plasma enables you to checkpoint the state on all those systems on Ethereum, on Layer-1, and they won’t be able to steal our digital assets because they’re easy to pull back on Ethereum.
Speaking of gaming, we at Hashed believe that blockchain-based gaming will kick off the mass adoption of blockchain. That’s why we launched our Hashed Labs program to support dApp and game developers. What do you think of and expect from blockchain-based gaming?
I agree with your thesis that it’s such a good fit, that game publishers for so long have essentially owned the digital assets and there have been lots of skins markets so it’s probably a multi-billion dollar economy. On Ethereum there’s a composite token standard, and there’s another standard that enables you to move non-fungible tokens between different platforms and so I think that’s going to drive a bit of a revolution in gaming. You can have an in-game economies with real assets and I think it’s just a match made in heaven.
So at the end of the last year, lots of people were talking about ConsenSys and its reorganization. And you called it ConsenSys 2.0 on many occasions. So can you talk about what exactly is ConsenSys 2.0 and how it’s going?
ConsenSys has reorganized itself many times since it was essentially one person or two people in 2014.and we noticed 18 months ago or two years ago that hey, there’s a real ecosystem, and it’s growing really quickly. Many different organizations, from big four accounting companies to IBM to many others, are rushing into the space and looking to compete in a next generation IT technology. So similar to the database wars, the OS wars, the browser wars, and the mobile wars and the game platform wars, the blockchain wars are happening. We’ll call them the blockchain collaborations rather than the wars, but there are so many opportunities now. And 18 months ago, ConsenSys started to organize itself to be more commercial, rather than just doing cool things. — and we’ve built a lot of successful pieces — we consider them building blocks, or core components, and we’re able to package them up now to offer a featured work suite or a tech stack or a fintech, DeFi set of tools, doing work in media, entertainment, journalism, advertisement technology supply chain, so it’s about getting more commercial.
To be more commercial, if you could select one industry or one sector, what would be ConsenSys’ biggest focus this year?
So we do quite a bit of work in a few different industries. So we do a lot of work in fintech, financial services, and increasingly in decentralized finance. We work with banks and central banks, and we have for quite some time. We do a lot of supply chain work. Our Viant project works with GlaxoSmithKline, an energy supermajor, a major American consumer products company, the World Food Programme, so lots of activity on that front. We’re also doing work with a very large luxury brand in Europe. And we’re in many niches. We’re in the energy niche, we’re trying to spin up a game studio, so we’re not very advanced on that front, but pretty excited about, as you are, the possibilities there. and of course, when you talk about different niches, I immediately think enterprise, but we also do the bulk of our work in the public blockchain space. So that’s projects like MetaMask, which is evolving — it’s about to release an incredibly cool mobile version of the client. It’s Infura, it’s uPort, it’s 3box, it’s Kaleido, which is a blockchain-as-a-service platform — for Ethereum it has about 2000 consortia on it right now. We’re really excited about the Pantheon client — we have about 85 engineers at ConsenSys and the bulk of them are in the PegaSys group, the protocol and engineering systems group. We have what we believe is the best client on the planet for Ethereum. It’s a microservices architecture, it’s Java and Apache 2.0, which means that it’s very easy for companies to adopt it. It conforms to the enterprise, Ethereum alliance enterprise standard, and most of the other clients have either licensing issues or other issues that make it less interesting for companies to adopt.
All right. So it looks like ConsenSys is working with many companies and existing organizations like central banks. I would think that it’s not very easy to work with existing big players or banks, especially when you pursue decentralization. Is there any friction, or —
Nope
None? Zero?
So blockchain is a next-generation database technology that enables us to build more trustworthy systems, so the trust characteristic comes from decentralization. It doesn’t have to be as maximally decentralized as public Ethereum in order to get benefit of increased trust in the system. So we’ve done work with the monetary authority in Singapore, the South African Reserve Bank, which wanted to build a real time gross settlement system which would link all of its member banks so that if something happened and the South African Reserve Bank computers were offline, the network still kept working.
So we can just take advantage of blockchain or the decentralization factor?
Yeah. Lots of companies are interested in some decentralization in their internal systems, and many consortia are very interested in decentralization because they can work, they can collaborate with the different companies in their space, or with the companies in their value chain, like a supply chain, and nobody has to own the machines, nobody has to own the software, everybody can essentially share ownership, share intellectual property, and it just makes it easier to build collaborative systems, whereas it was much more difficult before blockchain.
I think it’s very practical, but some decentralization idealists think that things should be fully decentralized, with applications and governance, and to do that, they think that the concept of Decentralized Autonomous Organization (DAO), is very important.
True. Decentralized organizations are very important. Again, we look at each different use case and build the appropriate architecture for each use case. Not every use case needs a blockchain, not every use case needs decentralization, and some use cases, I think, will need the maximal decentralization of the Ethereum platform — if you’ve got a token that’s very valuable, there are well-resourced entities around the world that are patient and are good at finding flaws in systems. And if a system is not sufficiently decentralized, it will have vulnerabilities they can take advantage of. Decentralized organizations — I prefer to call them DOs, instead of DAOs. Early on people started calling them DAOs, but they’re not autonomous, really; or the ones that we built are not autonomous. All the decentralized organizations that we built have people in governance. If we build a decentralized organization and it’s sufficiently well-architected so that it can run itself without governance by people, you sort of snip the strings, and maybe it offers a service to people as customers, but no people actually control that. So I would call that a Decentralized Autonomous Organization.
So I want to ask you about tokens. A token is an essential instrument to make a DO, right?
Sure. So any protocol-based open platform, if you’re trying to incentivize people to share their resources, or to participate collaboratively on a platform in different roles, you usually have to incentivize them in some way, and tokens are very ideal, a very focused mechanism to bring people to these protocol-based open platforms. So they can be a base-level one, like Ethereum or Bitcoin, or they can be something a protocol-based open platform built on Ethereum, like the Ujo music platform, or the Civil journalism platform, where people own tokens in order to stake their newsrooms on the Civil platform and pledge that they will be ethical, because it’s a platform for sustainable and ethical journalism. People can also use tokens to reward the journalists or to try to revise, improve the constitution that all the journalists are adhering to.
But even if a token is useful, I’m thinking of the big challenge last year for token value capture mechanisms. Do you have any thoughts on that?
I think we need to build good software, that people find compelling, and want to use, and then the token value capture will take care of itself. It’s not about value capture, in tokens, it’s about building good software that people will care about.
So you think that the problem is value creation, not value capture?
That’s well said.
That’s interesting. So do you think mechanisms like Bitcoin, medium of exchange, or store of value can still capture value from the network?
It has captured lots of value, hasn’t it?
It has, but the followers have failed with the same design and same model.
Sure. Well, systems have network effects and liquidity. If there aren’t, if there are two identical systems, and I find out about them at exactly the same time, and this one has exactly 99% of the people and the other one has exactly 1% of the people, this is going to be less valuable, it will have much less liquidity. I’ll be much less interested in it. So add to that the fact that this one started much earlier, and you have a winner, in the Bitcoin space anyway.
Because many people are disappointed in tokens’ value, or price, they’re now very skeptical about token sales. So what do you think about this?
I think people should build better software. And sell a token that makes a lot of sense on that open platform and nobody will be disappointed in that. A lot of people were disappointed with the user experience of the Internet, in 1992 or the web in 1993, and 1997, and software developers kept building better and better things, and 20 years later society has been profoundly transformed.
So you’re still very open to token sales, if the software’s valuable, right?
Yes. Don’t build terrible software and tokenize it.
All right. So the actual final question: do you have anything you want to share with our audience, or potential dApp developers?
Anything I would say would be very biased, I guess. An unbiased thing to say would be: learn the technology, build what you want to build, do something that you’re passionate about. We are at the start of maybe one of the most epic revolutions in history, where, essentially, the automated trust age is upon us and there will be so much to do and so much to create.
All right. That’s about it. Thank you for your time, and for the great interview.
