On September 20th at Consensus: Singapore 2018, NEO’s Da Hongfei participated in a twenty-minute panel titled “Ethereum meets NEO” with Ethereum Foundation’s Virgil Griffith, moderated by CoinDesk’s Nolan Bauerle. Below is a transcript of the discussion.


Nolan Bauerle, Director of Research at CoinDesk: Both of you have a really extensive developer community working on your platforms, and it’s one of the interesting parts about both projects. So, can you explain a bit about the developers that are working on these projects? What are they building? Who are these people? And what kind of traction have they got so far?

And we’re not talking about the core protocol developers…but second layer stuff. So, Virgil, how about you start.

Virgil Griffith, Special Projects Lead at Ethereum Foundation: Well…most people are building financial applications, certainly, I guess I think our first killer app was probably the ICO, and hopefully there will be more…there’s always Crypto Kitties.

Recently I’ve been seeing a lot of interest in an insurance applications….parametric insurance…and also fun things like [adult entertainment token] SpankChain.

Nolan Bauerle: And what about NEO? What kind of stuff are we seeing your developers working on? Is it different, or a similar type of communities?

Da Hongfei, Founder at NEO: I think it’s similar, yeah. We have ICOs, NEP-5 tokens, equivalent to ERC-20, and we also have decentralized exchange, the most famous one is called NEX, they’re doing pretty good, and also we are encouraging developers to do gaming, and STO, the secure token offering, I do think that decentralized exchange, gaming, and STO probably will be the next killer app.

Nolan Bauerle: Now let’s talk about the languages that the developers are using. So, Ethereum has its own native language. Is this a limitation to the platform? Is there an advantage that NEO has, to be able to use such dexteritous languages?

Griffith: You can look at it both ways. So, there’s a history of domain-specific languages. Let’s say you wanted entries from a database, it would be a lot easier in SQL than JavaScript or C. And we’d probably claim that blockchain development is sufficiently domain-specific that it’s more like SQL. And there’s other things like this, like R, for statistics, or even just MatLab, for Mathematica. And we could probably say that blockchain development is sufficiently distinct that you probably want to have a special-purpose language for it. You know, don’t get me wrong, that does increase the learning curve, yeah, it does, but sorry [laughs] We promise it’s good for you [laughs].

Nolan Bauerle: And what about with NEO? Were those decisions that you made originally to have it easy to access from preexisting development community?

Hongfei: I do think it’s in general an advantage to support more languages, or those mainstream languages. NEO currently, we have a completely different smart contract system. Currently we support C#, Go, Python, JavaScript, and a few other languages, in the future we plan to support those mainstream languages. And not that only we support those languages, but developers can also use their familiar ID [integrated development] environment to do coding, to do smart contracts, so I think we’re lucky to make the decision to support those languages.

Griffith: I mean certainly, I mean, a part of me, I mean, I’m with you, we had that exact same discussion in the [Ethereum] foundation,

Hongfei: Yeah, but anyway, I mean, I think Ethereum was started pretty earlier. I think the decision with Ethereum’s smart contract system was made in 2014 or something, and we started our smart contract system in 2015 or 2016, so we have–we can learn from Ethereum. But anyway, I think Ethereum is moving to Wasm [webassembly, a proposed virtual machine upgrade], right?

Griffith: Yeah, but that doesn’t change the language you write in. You’re not going to write in Wasm. Wasm’s going to switch out the EVM [Ethereum Virtual Machine].

Hongfei: Yeah.

Griffith: But for those of you who don’t know, Wasm’s going to be great, you know, it’ll make your lives easier.

Nolan Bauerle: Now moving to the discussion round, decentralization. NEO has some critics in terms of the level of decentralization that the network is aspiring to. What are some of the metrics that you guys will search for, in order to sort of satisfy the appetite there is for decentralization in the community?

Hongfei: I think the design philosophy of NEO and Ethereum is quite different. You can see on a spectrum, the most left side, censorship resistant and the most far right side, to be compatible, compliant with regulations, to be compatible with the current existing business world. So we are probably [towards the right side], Ethereum is probably [towards the left side], and I think it’s a balance, it’s a tradeoff, a different design philosophy.

Nolan Bauerle: Are we still at that point in the industry, where we don’t necessarily know what the right recipe is, what the right ingredients? We’re all basically making soup here, and we’re deciding which one….that people will enjoy the most?

Griffith: I mean, Chinese tastes in soup are a little different from the rest of the world.

Hongfei: I think, one of the reasons why we remain not that decentralized is we do think blockchain technology in general is still unmatured, not mature, so everything needs to change really quick. If everything is so decentralized, we see the block size debate over bitcoin, it will be a pain to upgrade the protocol.

Nolan Bauerle: So speaking of upgrading and changing, let’s say on the fly. Ethereum has started especially on the authentication level. If you’re just using Ethereum to move ETH around, of course, a private key is enough. And it’s a clean deployment of public key cryptography. It seems that PKI [public key infrastructure, needed to manage digital certificates] is built into the protocol level of NEO. How hard would it ever be to unwind that? When were those decisions made, and do you really have the flexibility to change the platform on the fly?

Hongfei: The decision to integrate PKI is, we need identity. We need [that] someone can issue a digital certificate to some other addresses to say, this address belongs to someone else. So that’s, it is because we want NEO to be used in the real business world.

Nolan Bauerle: So on the Ethereum end, how can Ethereum bridge that gap, still be used for that business end. Are your hands tied, do you have access to that world as well?

Griffith: Yeah, I, you know, I hadn’t really thought about this before I started prepping for this talk. And I can see legitimate arguments going either way. I think at Ethereum we would probably be inspired a little bit by the original Internet, which kind of intentionally kept things simple. And it was kind of anarchistic, you know, you can plug into it and start sending packets without any kind of authorization. And we would say that unless there’s some kind of really, really compelling reason, where you get a huge efficiency boost by putting in more control at the base layer, we’d probably say “Oh, well, let’s just replicate [the Internet].”

But no, I’m sympathetic to your [Da Hongfei’s] view. And I’m fully confident that there will be some kind of identity layer on Ethereum.

Nolan Bauerle: Now, that decision that was made also has implications for scalability. So NEO can handle a massive amount of transactions very quickly. How do you intend to deploy that, how important is that for the platform, and how limited will Ethereum be going forward with its current iteration?

Hongfei: I think it’s just like I said, it’s a tradeoff. If you want every node to process all the transactions, to join the consensus process, it will be slow. So the design decision is, we need efficiency. So we decided to go with our consensus mechanism, called PBFT [Practical Byzantine Fault Tolerance, as proposed at Massachusetts Institute of Technology in 1999, and adapted for the NEO blockchain by Erik Zhang].

So first, if you are a NEO token holder, you are part of the governance, so you can vote, who will be the consensus nodes. And then, those few selected consensus nodes reach consensus every 15 seconds. So it’s fast, and throughput, or the TPS, should be higher than those blockchains that is fully decentralized. So, it’s a tradeoff.

Nolan Bauerle: And for Ethereum, what advantages do you guys have with the way that you chose do do this? And what has to happen for you guys to reach the scales that will allow you to perform at a commercial/industrial scale?

Griffith: We might argue that if you’re clever about it, you could probably perform at a commercial/industrial scale right now. I guess we would like to say, I think at Ethereum, we’re a little bit more technology-driven rather than business-driven. Now this may just be a bias, but I think we all have this vision of this glorious sci-fi future. And it’s like well, if we can’t have the glorious sci-fi future, why are we doing this? So we at least want to give it a shot.

So we…until there’s some theorem, where it’s like “Nope, this really won’t work”, and I doubt this theorem exists, but if one did exist, then we would eventually scale back, but until we encounter a wall that hard, no, we’re going to stay decentralized, for y’all. And for the future.

For those who haven’t seen the scaling roadmap, state channels and Plasma should be out by the end of the year, while Casper and sharding…these are just unsolved problems in science. No one knows how to do these. No one, and, you know, it’ll probably be a hard fork or two away. Sorry.

We’ve had a problem, where we’ve been a little too successful. So we can’t mess up anymore, and now we can’t make a mistake. Before if we made a mistake we would just roll it back, and, you know–

Nolan Bauerle: –Like, two years ago?

Griffith: –Yeah, and we’d look kind of silly. But the world doesn’t end. Now, if Ethereum crashes, it’s pretty bad. So, when your thing can’t break, you have to quadruple-check everything. And like, I mean, it takes a while, bro. It takes a while.

Nolan Bauerle: Well, it’s like the whole implementation of cryptography: it’s either bulletproof, or it’s useless. And it’s not a video game that you’re building.

Griffith: It’s really annoying…as a contrast, I used to work with the TOR project [an online anonymity network], and they had this thing where they refused to use any cryptography that’s less than five or ten years old. They said, “Yeah, we just don’t trust, we just don’t understand it”. And at Ethereum, nope, we’re on the bleeding edge. And we’re going to try and deploy it. And yeah, no, trust me. You don’t want us to rush this. It’s, it’s for your own good. [laughs]

Nolan Bauerle: And on the NEO end, are there sort of cutting edge research and development that the project is working on, stuff that the world hasn’t seen before? Or are you happy to take different inventions and continue to tinker and put them into the protocol, and allow them to grow that way?

Hongfei: NEO has an investment firm, an investment arm, called NGC, NEO Global Capital. The performance in the last six months is really good [laughs], and we invested in one company called Starkware, they have a new zero-knowledge proof.

Griffith: And we gave them money too!

Hongfei: Yeah.

Griffith: Yeah! All right, Starkware! [Griffith high-fives Hongfei]

Hongfei: So, they have a new privacy-conserved scheme. And we also invested in Certik, they’re doing formal verification, to improve the security of smart contracts. And I also recieved an email this morning, we are about to invest in a project, they are doing quantum-resistant computing, so, yes, we are doing those intensive research.

Nolan Bauerle: So going back to Starkware, what is the appetite to incorporate those levels of privacy? We’ve heard that Ethereum is guided by it.

Griffith: No, no, we don’t do privacy, we do transparency.

Nolan Bauerle: Pardon me.

Griffith: Privacy is a whole different, it just makes your life harder.

Nolan Bauerle: Well then, on the [topic of] putting cryptography first, putting cutting-edge cryptographic tools first, is that something that is guiding NEO?

Hongfei: NEO probably wouldn’t be the first mover of those new cryptography schemes. We will wait and see how it works.

Griffith: Actually, I have a question, I wanted to ask about your research funding. So, I was talking with some of my friends about what we should talk about, and they said “Oh, we should find ways that we can work together”. So it sounds like, we’re already funding, we’re already joint funding some people. Maybe we should keep doing that.

Hongfei: What do you mean?

Griffith: So we’re both funding Starkware, for example. Maybe we can fund people to write papers, about some generic stuff.

Hongfei: Yeah, we should do that. Yes.

Griffith: Oh, okay, let’s do that! Well, that sounds nice! All right.

Nolan Bauerle: So there are ways, NEO and Ethereum, it’s pretty cool. [audience applause]

Griffith: One nice thing about this space is there’s still a lot of low-hanging fruit, and realistically, we could each give like a million dollars to some research group, and they could publish something novel! And it would benefit both parties, and it can be chain-neutral. And it’s great.

Nolan Bauerle: So any other chain could adopt it, and it would just be out there, as open-source?

Griffith: Well, I don’t know if it would be an open-source package, I think more likely it would be like a theorem or equation, saying like, “Hey, by the way, if you wanted to know the best peering rule, here it is”.

Nolan Bauerle: So moving on a bit to users on other platforms. We see of course that Ethereum, of course, is entrenched as a global technology. Is there a regional issue with NEO, is that okay? I mean, the population of China is incredible. Is it a regional play here, or is it also attempting to be a global player?

Hongfei: I always use the word that NEO is “started” from China. For the first 2 or 3 years, 99 percent of the code is from Erik Zhang, and then, the fire just ignited. And now, more than 80 percent of the code is from across the world. We have three big community developer groups. The biggest one is called City of Zion, most of their members are in North America and Europe. And we have NEO Research, they’re based in South America. And then we have NEL [NewEconoLabs], they’re based in China. So the development is, I would say, very, very global.

And those dApps, decentralized applications, they’re from everywhere in the world. Not just from China. I would say that one-fourth or one-third are from China, the others are from the other part of the world.

Nolan Bauerle: And what about node distribution or coin distribution? Do you guys have any insight as to what is going on there?

Hongfei: Yeah, because China cracked down on the centralized exchanges in China–

Griffith: –Sorry about that.

Hongfei: [Smiles]–I would say that most of the NEO tokens, the token holders, are outside of China.

Nolan Bauerle: And on Ethereum’s end, of course, you guys are absolutely everywhere. Except, is there a problem getting into China?

Griffith: Well, we were actually talking about doing more stuff in China, but I don’t know, they kind of made it clear that, they didn’t really want us there. And we’re a foreign company, maybe if we were Chinese, maybe we would have tried it, but–

Hongfei: –You should visit our office.

Griffith: –China’s like yeah, we really don’t–

Hongfei: –You should visit our office in Shanghai.

Griffith: –Oh, okay. Well, um…

Hongfei: –Please.

Griffith: –Yeah. Well, I mean, I’ve been to Shanghai a few times

Hongfei: –Yeah?

Griffith: –Yeah. I mean, I guess I can keep practicing my Mandarin. I mean, I don’t know, basically we, I mean Vitalik speaks Mandarin, and I mean we wanted to go to China, but the government is like, “We don’t take too kindly to your kind here”. And we were just like “Oh.” You know, and so we went to Singapore, and now we’re here. And if China changes their mind, then we’d be happy to come back.

Nolan Bauerle: And on your end, let’s say, I’m not sure how relations are with the powers that be in your country. But is there any way you could make introductions, and perhaps help Ethereum smooth things over, over there?

Griffith and Hongfei: [Laughter]

Hongfei: I can make introductions to all the blockchain technology companies, or projects. Yes, sure, yeah.

Nolan Bauerle: Open the door, and then maybe, Ethereum can go to China.

Hongfei: Actually, the Chinese regulators are pro-blockchain technology, but they’re against speculation, and raising funds targeting the general public in China. Retail investors.

Nolan Bauerle: So it remains that the killer use case, which is true, it is the first killer app of Ethereum definitely was the ERC-20 and ICOs, but that’s just not welcome.

Griffith: I mean, it would have been nice, if it was something–if that was not the first killer app. But, you know, you work with it.

Nolan Bauerle: But it was!

Griffith: Indeed.

Nolan Bauerle: But it was.

Hongfei: Targeting, raising funds, not targeting accredited investors in China is not welcome, yes.

Nolan Bauerle: What about Ethereum seeing these non-fungible tokens and CryptoKitties, are those things being developed on your platform?

Hongfei: Yes.

Nolan Bauerle: And is that the type of thing that would be welcome, from any platform?

Hongfei: Currently there are three competing proposals for non-fungible tokens on NEO. The community is still discussing which one is the better one to implement. I can’t speak for the regulators in China, whether it’s welcome, or not welcome.

Nolan Bauerle: And now, we still don’t know, I guess the size of what the world can handle in blockchain. Is there enough room for two platforms like your own?

Griffith: Yeah, there should be.

Hongfei: Definitely.

Griffith: I mean, even in things where it’s very much a winner take all, like in search. There’s still Baidu. There’s still Google. So, yeah. I imagine we might have to figure out how to do some kind of interoperability layer at some point.

Hongfei: Yes, sure…[inaudible]. There are two ways to do it, we can scale up. We can have layer one, layer two, even layer three. And we can scale out. We can connect with different blockchains, so they do their own business, but somehow they can talk to each other.

Nolan Bauerle: Well, really fascinating stuff, guys, but unfortunately, our time is not scalable. So, please thank our great panelists today.


Video of the above discussion can be found at the following link:
https://youtu.be/z98Sh3ldRNk?t=23310

Virgil Griffith helps lead Special Projects at the Ethereum Foundation.

Da Hongfei is founder of NEO and Onchain.

Nolan Bauerle is the Director of Research at CoinDesk.