Blockchain’s Scalability Conundrum: Dogecoin, TrueBit and What it Means to be a State Channel
Blockchain’s scalability problem has been the elephant in the room throughout its development. Transaction times are slow and the energy consumption is huge – fundamental problems that prevent the technology being used for genuinely workable use cases and achieving anything resembling mass adoption.
Without effective solutions, blockchain will never be capable of catering for transfer volumes comparable to existing systems, so it’s a problem the community is actively pushing to solve. One such solution is TrueBit, which works with Ethereum smart contracts to build a verification layer for the blockchain.
TrueBit is the vision of Jason Teutsch, a mathematician and academic who has held postdoctoral positions at the National University of Singapore, Penn State and Universität Heidelberg, as well as research positions at think tanks RAND and IDA along with multiple Fulbright fellowships. The organisation is working on a number of projects that address elements of the scaling conundrum.
Patrick McCorry is an assistant professor at King’s College London and was the UK’s first cryptocurrency PhD graduate. He is directly involved in countless blockchain and cryptocurrency projects and is working on practical, deployable state channels as a potential scaling solution.
Both Jason and Patrick spoke at ‘Off the Chain’, our master workshop which took place in Berlin earlier this summer. Across the two-day event we saw over 25 speakers present their work and their ideas about how best to utilise off-chain solutions to scale blockchain.
Patrick sat down with Jason at the event to discuss TrueBit, scaling in blockchain, the semantics of the industry and blockchain’s foray into the world of art.
The Exciting Thing About Blockchain
Patrick McCorry: Hi Jason, thank you for accepting this interview. I think the first question I should ask you is that: you were an academic for over 10 years, you wrote lots of papers, you proposed lots of protocols, but actually TrueBit is the first one where you thought ‘I want to go and build this’. So what is that transition like, between academia and industry? And how do you find the funding situation?
Jason Teutsch: I think maybe it’s the case that TrueBit was the first one that could actually be built. When you’re writing theorems in computability theory and you show some impossibility result, well, that kind of rules out the possibility of building it. I still have a soft spot for algorithmic randomness. The exciting thing about blockchain is that there is a lot of low hanging fruit and you have an opportunity to get in on the ground level.
On top of that, there are funding opportunities. You don’t have to wait a year to see if your National Science Foundation (NSF) grant comes in. There are a lot of economic opportunities in this space.
The problem with academia is, though it’s fun, you can write one paper and write another paper, but if you actually want to build something you kind of have to throw yourself out there.
We put ourselves 100% into TrueBit and I’m happy that we actually have a working prototype. We’re getting our first product version out there with Livepeer and some other projects, so we’re exciting. There’s a lot of people who want to build, so I guess there’s the fun of seeing the thing out there and working in the real world.
Also, with cryptocurrency being interdisciplinary – I like being at the interface and talking about not just computer science and mathematics but economics, psychological, sociological, business... Law, in particular, is interesting – there are a lot of questions around the philosophy of law that are really interesting
Does Anyone In The Blockchain Speak The Same Language?
PM: One problem I always find as an academic is that when I talk to someone else in the industry, we always have different terminology. Thankfully for me, I’m not trying to convince them to build something. Do you have the same problem? How do you find it when you do talk to engineers?
JT: Does anyone in the blockchain speak the same language? We have a lot of words, but there isn’t really a consensus on what the semantics of our language is. I think that’s something that should be refined over time. Even ‘state channels’ – is TrueBit a state channel? Well, if we knew what a state channel is I could answer that question.
I took the safe route and I didn’t talk about TrueBit at the talk today. I think Doge Ethereum is safely a state channel. Patrick, what do you think is a state channel?
PM: I can answer that, I guess! Throughout this workshop we’ve had maybe five or six different channel constructions explained, and everybody has a different definition – I tried to highlight that in my talk as well.
For me, a state channel is when you have N parties, and they all want to agree to a state update. So actually what they’re doing in this local consensus protocol is they run an application or a smart contract.
The whole point of a state channels is, one: there’s collusion among everyone else and you’re still safe and you get your coins back. Two: if one person doesn’t cooperate, you can go to this global blockchain, which is run on a global consensus protocol, and that’s your route of trust. It always guarantees safety for you, so people can’t cheat you and you get your coins.
The Doge Ethereum Bridge
PM: I want to ask you about your Doge Ethereum bridge. Is it a state channel? I know for the first part of the bridge, I believe it was Doge going to Ethereum, you guys won a bounty for that, and what I thought was really cool is that you guys donated that to the art project. How are you finding the second half of that challenge, going from Ethereum to Doge?
JT: First of all, I think it’s important that we have some projects that test the limit of what is a state channel. I grew up in the world of mathematical logic, where we ask what are the limits of the language itself, so testing the limits is interesting.
There was a bounty for Doge Ethereum - there still is. We took part of the bounty for building part of the bridge, and it’s still under construction and I’m hoping that some of the ideas I put today are going to influence our design decisions. And, of course, the physical manifestation of that bridge, which is our art project, is basically fully specified at this point in terms of the engineering design. We have a site for it, which is going to be in Vancouver, so it’s very exciting.
I said originally that the reason we started building the bridge was purely politics. We needed to get the miners’ attention so that they would install the software. Now that we have this new construction, we can actually build the bridge without a fork but still enjoy the art and have a place where people can learn about TrueBit technology, do something with cryptocurrency that engages in a different way.
So, it’s about education and about giving back to the community, which has been an important theme for our organisation in terms of our different charitable research art initiatives.
PM: I love the crypto art – I saw it when I was in New York, it’s awesome.
JT: Ask not what blockchain can do for art, but what art can do for the blockchain.
PM: Let’s go into the detail of the second half of this project. How can I prove in Doge that there have been blocks in Ethereum and I want to unlock my coins? So, the idea was that we go from Doge and we log coins into Ethereum. These become Wow tokens.
Then you want to lock or burn your Wow tokens and go back to Doge. How are you finding that challenge? That’s really difficult, because Doge is not made to do this, a bit like Bitcoin.
JT: There are several challenges to building this bridge. You want to somehow relay information from Dogecoin into Ethereum, and that is a non-trivial thing to do, because when the information comes in how do you know whether the information coming in is from the real Dogecoin blockchain or whether its coming from an orphan block.
Number two is: how do you know if it’s a real block at all? You have to somehow check the proof of work in the Dogecoin, which is why we started TrueBit in the first place – so that we could check the proof of work.
It costs too much gas for what you can do with a smart contract in Ethereum, so the whole thing was a workaround and we built this new off-chain architecture and that’s the TrueBit protocol. Then, of course, you want to do it efficiently without the gas.
Then the other major challenge, is that there’s no ‘lock’ in Dogecoin. A political solution is to say, let’s add something new into the vocabulary and get everyone to form a consensus over what that means.
The other option, which I presented today, is the crypto-economic option, which was to say that we create economic collateral or incentives so that someone can open a bridge by putting in a deposit. So, you put money in on the Ethereum side, I send you money on the Doge side – if you, for whatever reason, move your collateral on Eth, whoever is holding the Wow tokens will grab your stuff. So this is the basis of the collateralised construction.
Who Doesn’t Love Doge
PM: So I’ll ask one last question. There are two parts to it. So one: What applications of TrueBit are you really excited about? There are lots of companies that must be talking to you now, what do you think is going to come out of that? What are you excited about?
And two: What are you excited for in the space? What makes you wake up in the morning and, instead of thinking of Doge, you think ‘I love this area’.
JT: Yeah, who doesn’t love Doge. So, I think that maybe they are also one and the same. The vision of TrueBit is to allow us to do computation on the blockchain, and the other piece that I think is really missing is storage.
The whole story here is about building autonomous machines – Bitcoin is an autonomous system, Ethereum is another autonomous system, but if we really want to build powerful machines we need a machine that has access to resources and it can do compute and it can do storage like a regular computer, then you’re opening up new possibilities. I think that’s exciting.
In terms of TrueBit applications, there are many. Livepeer I mentioned, which is decentralised live streaming videos. Any time you’re running out of gas and want to do the computation on chain you can use TrueBit. You can also use it, for example, to check bullet proofs. So you can get some zero knowledge actions on the chain. People use them for all kinds of things – voting is another one.
This interview took place at the Binary District ‘Off the Chain’ master workshop in Berlin. If you’re interested in more blockchain talks and interviews like this, we have a variety of events coming up over the coming months, bringing together leading developers, researchers and entrepreneurs to discuss the most pressing issues holding back emerging technologies from reaching their full potential.
View the full schedule here.
Illustrations by Kseniya Forbender
To contact the editor responsible for this story:
Margarita Khartanovich at [email protected]
- Artificial Intelligence Isn’t Ready to Take Over From Doctors and Nurses, Just Yet
- Machine Learning Vs. Analysts: Will AI Eventually Replace Data Scientists?
- Why The Danger of Deepfakes Is No Danger At All
- Sarcastic Robots? How Deep Convolutional Neural Networks Are Making AI Worryingly Human
- Traffic Lights in The Sky: Flying Cars Can Appear Sooner Than You Think