The Future of DLT with Brian Behlendorf (Hyperledger), Jutta Steiner (Parity Technologies) and Gert Sylvest (Tradeshift) | Disrupt SF (Day 2)
3:01AM Sep 7, 2018
From from Parity Technologies Jutta Steiner and from Hyperledger, Brian Behlendorf are here to discuss the future of distributed ledger technologies, open source, how it's going to fit into and dovetail with the crypto revolution, and to unpack that subject with them is senior writer based out of gay Paris is Romain Dillet. Round of applause, everybody.
Was hanging around them, wanted to come on the panel, I guess. So let's start. We're going to talk and there's not many of you, so don't know if you're familiar with blockchain stuff or not. Do you guys familiar with Hyperledger with Parity? Not really. Maybe. So maybe let's start with you, Brian. You've been working in tech for decades now. Right?
I'm old. Yes.
And on many important or port open source projects that power many of the important bricks of the internet and tech in general. But for the past couple of years, you've been working on Hyperledger for the Linux Foundation. Can you tell us in a few words what it is?
Sure. So like other Linux Foundation projects, it's an attempt to pull together a network of hundreds of different companies and thousands of developers, both with those companies and beyond, to build a collection of building blocks for blockchain technology. And as we understand it that means distributed ledgers smart contracts and all the sort of things that go along with that and to try to say, you know, there's a lot of great ideas out there about how these things should be built.
But ultimately, plumbing is kind of boring, right? Like plumbing we should spend some time on, we should try to make sure that it's right. But really, as companies in this space, we should be trying to build things on top, right? And just like with operating systems, and container technology like we do with Kubernetes and all these other domains, if we can just work together on enough of these common building blocks, then maybe we can kind of spend more of our time and effort on kind of the interesting apps on top.
Yeah, and I think the Linux Foundation has been key when it comes to building these kind of building blocks for the rest of technology, so makes sense. Jutta, you've been working on Parity projects, and you have various different projects. Can you can you tell us about maybe some of the most famous one as like Polkadot and Substrate and how they work together.
So we started early on with like, a completely new implementation of Ethereum when you started the company, and then over time as we learn more how things are used through the different things we build, sort of this idea of, of Polkadot emerging initially as a tool for making blockchains and tropical. But then over time, it evolved like as we learn, like how what the efficiency of of blockchain are like governance is one big thing that we're working on. So Polkadot comes, it's an it's a new blockchain framework to allow people in particular to innovate really easily and adopt other technologies as they come along.
So a framework that we have abstracted into a toolkit called Substrate, which is a general framework where it's now becoming really easy for people to launch their own chain, like, think of your app as a chain or features, a chain, like that's that's how we think things, as a box, things are evolving, and we want to make what we've learned in the past available through through that through packaging it up like the networking stuff was really difficult.
Like that's that all comes bundled. So all you need to care about is how you basically customize the state machine to your needs, like whether you limit the subset of instructions because it's all you need, or the way how you need to do governance because you're like a supply chain application for example, and there's certain things you need to comply with. So so that's what we are we're pushing towards with Substrate and Polkadot.
And Gert, you basically on the other side of the equation, you've been working at Tradeshift which is a pretty well known company focused more on supply, supply chain financing, and in all the various ways possible. And at some point, I don't know when it was last year, something like that, you realize, okay, maybe we should think about building some blockchain projects because otherwise we're going to get disrupted. How did you came up to this realization?
So I think it was maybe not so much the disruption piece, but I think it was more actually the alignment between what we were doing. So, you know, we founded the company in 2010. And what we do is really digitizing supply chains. And it was based on the notion that, you know, how people are connected and in their private lives, and social networks and so on. This has not at all happened for in the business world. So there's very little about business to business interactions that is digitized in any way.
So we basically founded the company of a vision that it should be possible to break down all the silos that exist between businesses. And and I think blockchain is kind of holds some of these promises, it's the fundamental idea of having a ledger that is shared between multiple participants, whereas the way that that companies and supply chains everywhere in the world operates today is by hoarding data in silos and exchanging, you know, very narrow targeted pieces of messages between them in order to synchronize their worldview, and I think the social economic impact on missing out on on actually digitizing and creating liquidity for this kind of information is just enormous.
I mean it's $9 trillion at any point in time standing out between buyers and suppliers. So a lack of liquidity that especially hits the small and medium sized segment of companies, so so that's when we built for each of those basically the idea of could you invite the finances the logistics parties, the sellers and the buyers into the same transactions and what is the value you could derive by doing that? So, obviously, we need to be in on blockchain because you know, it's a Hollywood playbook of most b2b blockchain cases. That is, what would happen if we actually invited 2, 3, 4 parties into the transaction? What is the value we could gain by doing this?
And talking about plumbing, give us an update on your project. What have you been using, are you doing everything in house, or leveraging other projects?
Yeah, so so I'm heading up a unit in Tradeshift called Frontiers, and the way we are working is basically exploring emerging technologies right now, spending most of my cycles on blockchain, but we want to develop it through partnerships creating joint ventures, deep partnerships, and spin off companies.
And some of the areas we're working on right now is biggest one is is financing to small and medium sized companies. So basically a tokenizing off I owe us between buyers and sellers and making those available on the blockchain. So think of of the blockchain as a marketplace for these kind of things. But with a level of transparency you haven't seen in these kind of instruments before, apart from that because we are supply chain and it's a you know, food safety and transparency in supply chains.
And we also have a track where we look at basic function of how do we do audits. Can you move that process from being a post things have happened and you basically fixing exceptions to preventing exceptions and and try to do audits on a transaction basis and and turn it into a valuable thing instead of a chore.
Okay, um, let's let's dive deep dive a bit and find out like what what makes you different? Like, if I look at Hyperledger and Parity Technologies, how would you say you different and how would you differ in your approach to kind of the same problem.
I and and I mean, I won't to speak for for you that I think one important difference to understand is kind of the distinction between being an open source project and the nonprofit, kind of building these underlying layers, but we're not going to go in and build a network for you. We're not going to go and build an app for you. You can't really hire us to do that. Our business model is membership, right? Through the consortium, right? And it's very much like, you know, I think parity could be like the Amazon Web Services of the space if they want.
I don't know if maybe that's the metaphor you would choose, but like, you know, building on top of like these underlying technologies or writing writing their own, and we have a lot of our members you know, we have a vendor directory with 60 different companies that are building products and services on
Many of them are very big, like IBM.
Some of them are big, but some of them are startups in fact, down in blockchain alley, there's a couple of them there's one called Scroll Network that is has just a real simple kind of file sharing kind of application design for large files but using Hyperledger Fabric sorry I'm sorry, Hyperledger Burrow which is actually our Ethereum VM implementation as the kind of checksum and and an audit trail for file sharing. And it's a really fun small company. In fact, most of our members at this point are smaller midsize businesses. So we have the IBMs and the Intels and the SAPs and they are building businesses on top of this.
We also do have a lot of international companies, about a 20% of our members are from China and another 20% from other parts of Asia. So, but we know that part of our mission is to lift up these companies, help them build businesses on top of this underlying plumbing as much as it is about being the neutral home where that software is built.
Let's get another opinion on that. And ask Jutta, if you saw a company that say, let's say a small company, a medium company, a big company, would you recommend them to go to Parity or take Hyperledger's work and use it for their own stuff?
I guess it depends, like what what you want to do at this stage. I mean, the reason why we started the company back then was like, we saw an opportunity for taking blockchain and building it into the future stack of the of the web where we decentralize wherever we can decentralize to make things more secure, to, to put ownership into into the users hands wherever possible, and so and so this is what what we try to do and it's a it's a lot of work. I mean, and that's not to be underestimated. So we have the ambition of being the, the technological leader in the space and, and building out this infrastructure.
So we're really trying trying to push to push the forefront of, of what's possible. So if you're keen on exploring that, then I think, you know, you're right,
I'm launching an ICO. I want to create a new and new Ethereum based token.
I mean if you want to do that like everybody is using I guess at this stage our software because it runs the Ethereum network. There's stuff we've done to help the community achieve a better way of doing ICOs. There was a service we build called pickups where we tried to build an open service for doing passport verification to help, like, to hit the sort of minimum standards that you should have in an ICO so these are the things we're trying to do to actually build the different layers and and and pillars for this decentral web that many people hope is a better way of building businesses going forward.
And you stand, from your standpoint, have you seen as sort of seasonality. And and of course, there was a huge wave at some point at the end of last year. How do you see the trend right now? Like, do you see a lot of projects that come to you?
I'm pretty excited at the moment to see how things are slowly moving actually to an a building like building out phase. If you look at the different products like we've we've cooperated with like a few select projects to actually explore how how things can be used, like one was the the, the project we did with the workshop program where we deployed on a pretty large scale, I think the larger scale on the biggest projects or far like the public network technology we built for Ethereum in a private setting.
And that proved to be like a really useful thing because like there were no bugs like that we that we found in the deployment, right. The only thing we found was like an issue with the iris scanner, but like having her the texts that were promoting on a public network proved so beneficial to the systems and and that's that's where I see I see things moving forward and I think like even though like it's all like huge scandals that we've seen around Cambridge Analytica and Facebook, like I think it has helped to see what the value of decentralization is beyond like early use cases as ICOs and whatnot like that.
We actually need a better infrastructure that's fundamentally different so that it can able people holding your data and whatnot and that's what we want to want to build a more people approach us to the like, few select use cases and show how things can actually work going forward.
And at Tradeshift. Do you think, because we talked about the governance a bit earlier in the conversation, do you think by building a blockchain there's a risk that you see other actors that are going to rely on your blockchain and maybe take over some you have some of your clients?
There's always risk as someone takes over some of your clients. But but I actually think we are pretty perfectly positioned to, to actually create real value from the blockchain. And I think it's a common thread that when you look at a lot of the business to business blockchain use cases and also some of the use cases that companies have invested in that they are almost more about economics than they are about technology, right?
So if you're looking at, for example, a project that looks at creating transparency in the supply chain, you know that you need to onboard a lot of different actors, small companies, large companies across many different regions of the world. So the question quickly becomes, how do we actually incentivize people to join into these networks? And I think that's one of the promises of blockchain, that is, you can kind of connect the thing about what is the information we share, what are the contracts that govern this information and how is it connected to the final settlement of value between participants?
And that's actually an economic question, right? So we have this technology that promises to enable fundamentally new economic models. And I think that's a great thing that that companies are starting to asking those questions, you know, what happens if we start acting a little bit like some of the digital leaders in this world, which are all two sided market models when we start inviting multiple parties to the transaction and and try to apply that in into into the supply chains, for example. But I think what companies may be under estimated sometimes,
You know, it's building the blockchain Castle in the desert thing that you have all this nice technology, but you still need people to agree on what are the digital standards we use. How do we agree on processes and rule books, what are the contracts we are governed by, what's the governance of the total infrastructure, who can opt in, who can opt out, how do we trust the information and so on and so forth, last mile integrations, user interfaces, and I think we pretty much in our business have all of this.
So I see the step for us to go into the blockchain and enable the 1.5 million companies in our network to actually join into the blockchain and pump volume into the blockchain, I think is a very, very small step compared to going the other way around. So for us, I see it as a great opportunity. Yeah, and as well as for the companies in the network.
Yeah, and it feels like it's sort of leveling the playing field as well. But maybe at Hyperledger because working on open source projects, do you think there's a risk that by building these open source bricks, some tech giants are going to take those projects and, and become this sort of sort of leaders of the blockchain world and it's going to be the same as tech as it is today with now on the blockchain?
Well this was the debate even in the earliest days of open source software, like, what would happen if Microsoft picked up the Apache Web Server and put it into you know Windows NT and nobody came
to contribute contribute
Exactly. Microsoft joined the Linux Foundation, you know, like last October and an internal thread went around which was basically, have we won, is it time to go home, right? But but but back when Apache talked about it and it was like you know, we have this license the Apache License that allows companies of big companies, small companies to take this technology and even make modifications to it and keep that to themselves and use it inside a very proprietary products, is that them stealing from us, is that them you know, like playing unfair and the the rationalization we came to is a if they picked up Apache and we're confident in our implementation of the HTTP protocol, then what we're doing is helping the rest of the world pick up that standard, right I mean, because most most of us were not motivated by an all software must be free thing so much as we are by, we want to protect the web, make the web an open place that didn't become the Microsoft web, which we kind of ran the risk of becoming in the late 90s or later on the Netscape web, you know, optimized for this, optimized for that, but that it was instead this open playing field.
And the fastest way to get that a standardized adoption we felt was by having a high quality production quality reference implementation and very freely allowing these companies to embed it and the gambit was, it's a numbers game, the more organizations that adopt it, the more organizations will find a bug and rather than fix the bug and keep the fix for themselves, they'll contribute it upstream. That's so much more efficient, right, and so much more, you get value out of it and turn out that paid off, right. I mean,
everybody gets value
and, and even though IBM did, like, adopt Apache into their IBM web sphere, right, you know, they turned around and contributed a lot of code back into the Apache web server and ported it to Windows, right? I and and I think we found a balancing act in the whole of the open source community where as long as the the code is under an open license, right, but also the development process is public.
That's the key thing. Like, as long as there's the decisions aren't being made in the backroom, serving the agenda of just one of the companies as long as it's a do-ocracy and if you want to influence the direction of the code show up and say, I'm going to work on this feature, and if anyone else wants to help me great, but we're going to get this upstream. That's how stuff is done, that's how agendas are set in the open source world, and that gives us a layer of insulation against I believe against being co opted to serve just the needs of one entity.
And this one advantage and open source software as well it's it's pretty good when it comes to bugs because we have a lot of contributors looking at bug and and forking the code to to make a patch and stuff like that.
You can't depend upon I mean there's there's certainly you know it with enough eyeballs all bugs are shallow as a principle, but sometimes there's not enough eyeballs so you do need something to you need you to invest in growing the community, you do need to invest and making sure that that pipeline from being just an ordinary user to being a bug fixer to being a contributor to becoming a core maintainer, like the activation energy for each of those hops is as minimal as possible, right?
And so that's, again, part of the science that we're trying to develop at the Linux Foundation as a whole, and the science and art really, of community management and ecosystem management to try to get that that that flow of developers into the code.
And what's your community strategy at Parity when it comes to open source projects and stuff like that.
So, I mean, we are we are open source for reason. And that ties back to what you said like I mean, you want to have a strong community around like your code base and and people adopting it. I mean, really, what we've what we've built, I mean, there's there just needs so much building at this stage that developing stuff in the in a closed way, like you won't get anywhere. I mean, you need you need a community around it and, and being able to share stuff like also to just learn there's so much learning that needs to happen at this stage like governance, you can't do this in isolation.
Like all the things we've seen now with the theorem, all the problems like, it's only possible to improve that and and build those future systems if you have that community.
And at a Tradeshift are you still at the learning point as well or are you
In terms of community?
Not just community, but blockchain in general.
I think to be honest I think everyone is. Yeah. So so I think the technology is developing so fast so the only way to really stay in the loop is being part of it, and actually learning by building things and and try to push those are children out into the real world.
Would you consider yourself as a researcher
Not in the academic sense of the world no. We are definitely very business focused but we are definitely on a journey of exploration or research if you can call it that, but it's really with a with a fundamentally commercial outlook right? So, so we don't want to be the guys that run around and in lab coats and and have, you know cool blockchain laboratory experiments, we want to make a lot of experiments so we can find the 3, 5, 6 things that can actually take on their own commercial life and make a real impact.
So what we're driven by in my organization that is really things that at a fundamental level either has an economic impact on for companies and the network and and, you know, we are moving half a trillion dollars here in 2018, which is three point something percent of the global economy and, and whatever comes out of this lab has to register on that scale. So that's and that means we need to commercialize these things and show that it can deliver value to to actually complete.
We just have a few seconds left. I'm going to ask you three one question. If you had to say when are we going to hit version 1.0 of Hyperledger, Parity and Tradeshift blockchain projects? When would that be?
We'll hit version 1.3 of Fabric about in a month and across the board I think we're very much at a at a time when you can put it into production now. Fabric and Sawtooth and Burrow and then these other technologies, we're ready.
So with Substrate we are moving now to to beta phase where people and we have a bunch of projects using it now building their own chains on top and with Polkadot we hope to be in a state where we can launch the network end of next year.
And at Tradeshift?
Yeah, 12 months from now. We will definitely be moving significant volume on the blockchain and every, no technology stands alone so where it has rough edges we and everybody else need need to make up.
Yeah, it feels like in blockchain three months equals one year basically. All right. Thank you very much.