📜

Talk Transcript

Stephen Tse: Uniform Scaling on Shard 1 & Economics Games on .1.country

Hi everyone, I hope you guys liked the video we just made. It was from a few months ago but still represents what we feel like to tell you the message today, keep building. Keep being the builders and bring out the adoption, whether Web3, whether database, whether Silicon Valley innovation, that's why we're here.

So today there are two topics I want to talk about. The first one is uniform sharding, uniform scaling. The second thing is economic games. We feel like its not just about the infrastructure, but how to bring adoption is still the topic. I would say starting from Silicon Valley, where the next few years will still be the hardest thing and the best thing to build coming from hopefully this group of people coming together.

So in terms of uniform scaling why do we still care about building a base protocol and the blockchain to think about the scalability. We must ask when there are no more volume, no more transactions, no more users, why people still need infrastructure is what Silicon Valley is great for,  and the room full of builders that understand that it's always, always the infrastructure builder, the innovation, the research that paves the foundation of the future.

Even though it has been a great peak earlier this year, we still actually think the demand is much more to come. It's still only at 1% adoption for most of the possible products that everyone has already prototyped and building.

And our approach of using sharding has been a few years in the making and I want to tell you a fuller story in the context of different other approaches now. As you guys know, doing scalability has been the story for the last few years and I would argue for a few more years, if not the next decade to come. We only are serving 10 million users users using different DeFi and NFT products, maybe at best a 100 million users have ever had any tokens.

To be serving the billion-users level whether many of the social media, social network and many of the engagement that you can do to peer to peer in different communities. We think that there is no way the infrastructure can satisfy today. But there are a few more technologies now; layer two, app-chain, if not even zero knowledge proofs, have been the new approach to scaling, to do scaling, and solving the trilemma  of decentralization, security and scalability.

So today, I just want to touch a little bit on why sharding is still an important approach to this and then the second half of the talk, I'll go through why there are still many ideas for adoption that you can think of.

So the two things today is, hopefully you are just a crazy student thinking about Web3 is still exciting because what the technology is. The second thing is you are thinking of the user adoption value that you want to create for users but what are the implications, if not all this hype, and all these marketing campaigns, that people are lost in all the scams and like loss in value, that this is a great time to not just build technology, but bring it to the market. Marry the technology with the culture so that many more millions of users can do.

The first topic is the layer two. As you guys know, finally, only in the last few months Optimism and Arbitrum launched.

We think that there's no way that this is the end of story. We do know that Arbitum and optimism take more than two years to be build, starting from a side chain and now layer two in just a general infrastructure that anything on chain putting the data in transition for you to verify on Ethereum sounds like a great story.

Guess what? The story is only one year old. And it's already helping many of the DeFi projects, saying that we don't need another layer one protocol. Sounds like a great story, what can we learn?

We are not a base protocol, there are a few more base protocols launching every month, still. Where do, not just Harmony stands, but where are all the story of transactions? Where are all the pieces tied together? We all need a framework to do it.

I put it in the triangle so that everyone can easily visualize. As I said, let's think about scalability at the top we call it a uniform scaling story. But first layer we zoom in is a layer two. Layer two, we think the top project is Arbitrum and some of the few project here that you will hear about.

So in terms of optimism, the key problem they have is the sequencer that they do think about, not just proving it but actually putting a lot of transaction, both in sequence being a single bottleneck. And how does Harmony if not not sharding compared is we already launched multiple nodes, thousands of them in multiple shards even four years ago with the design of sharding and launched the mainnet three years ago to do  all this distributed processing. The key one is they can do layer two very well now in terms of verification, security, because that's Ethereum is designed for.

Layer two is very cheap now compared to, comparable to other protocol, than base protocols elsewhere. But they have not solved the single bottleneck problem if not even decentralization of the sequencer.

So I want to here to help you contrast of the approach that we take that we are able to grow linearly, I would say even boundlessly with the demand of the transaction and that's a contrasts on what a sequencer approach with a layer two that they still need to figure it out for both Optimism and Arbitrum, how to actually decentralize and paralyze the sequencer.

The second interesting thing we talk about is a fraud proofs obviously that they have done. Why fraud proofs is interesting is they finally figured out that you do not need every single data for you to verify they can randomly  sample it, not only like linearly, they actually have a two-dimensional sampling that become extremely efficient.

But the opposite of it of fraud proof is optimistic. They can only confirm this transaction optimistically and say you can keep challenging it for a couple more days. You need to watch the transitions, you may be rolled back if someone challenges a fraud, that become very problematic not just for large value transactions, but for actually the user experience, right? You can do all this arbitrage, you can actually guarantee the fund but how do you even design product? We think it's a fantastic use case now but that leaves much room for us to improve. Obviously for us, our selling point is two second finality, that's it.

We tried to push to 1 second finality, it's not as easy as you think, but it's finality to begin with. Finality of any transaction, any value, any volume, any TPS has been our breakthrough since four years ago once we take the sharding approach. And not only that, we have sub-sectioned, even the leadership rotation within the Shard so that if there's any lifeness problem, from any misbehavior within the Shard, we'll be slashing them because they put up proof of the stakes as approved to be part of the committee.

That we have done that for a few years to understand what Sharding allow us to do something called a view change protocol. For those that are coming from traditional database, if not distributed system, it has been the best research that we pick up from not just in the blockchain community but been pre approved in distributed system, database system, that we know we are bringing the best research to production.

I'm very happy to say that the last few years has been lots and lots of production issues but has been very very robust approach that we have taken, we cannot get it wrong. That case of research, many many best papers has been taking this approach.

That's why we think the sharding but also the committee approach of something called the pBFT, the Byzantine Fault Tolerance protocol to the maximum that not only that you can work in the back end, but can be working openly with thousands of nodes that can be malicious, that can have lifeness and view-change in that way.

The last one is more exciting called account abstraction that many protocol is talking about now and only possible now in layer two. Because to do any change in the Ethereum base protocol, they move as fast as possible already and that has been amazing moving from proof-of-work work to proof-of-stake. No-one, half of the world still thougt it's not possible, but adding features is very different.

So as a base protocol, we need to be far more progressive in not just really having done sharding and proof-of-stake for years now, but thinking about many of the features that is coming that would support smart contract wallet, that would support gas fees, forwarder processes.

So many of the things that we highlight, some of the top projects like the protocol NEAR because we are learning from the best innovation experiment in this space as well. And later on, I'll talk a little bit more about smart contract wallet as well.

The second best approach to the scalability problem is called the app chain and we think Cosmos has the best ecosystem and the approach to it that I've been doing for five years already.

As you guys know, Celestia is doing the data and even Binance and back in time Luna on Cosmos has really proven out the Tendermint consensus really works. The question is, does the IBC work? Does it interconnecting the consensus among all these action work?

We think so.

We think that currently there are about 200 Cosmos on Tendermint collecting, probably tens of them,  connecting IBC will be the one of the biggest ecosystem, if not even bigger than Ethereum potentially but why are we taking a different approach instead of app-chain  is why I’m here to tell you today.

App chain works, layer two works, what else? It's the uniform scaling. There's only one message to take home. One word will be sharding. Two words will be, uniform scalability.

App chain is great, except you have to run it - your own validator, your own token economy. It's like you getting many of the server and you bring it back to your garage and saying, "I'm loading up another Linux that have a cluster configuration, et cetera."

App chains is gonna be great if you are that. So we're taking much more of a, as you guys know, elastic cloud computing approach that any more demand that you have, you don't need to think about more validator or another token to secure your sidechain or app chain to do that.

But in that case, we must learn from the best of the approach. In this case, the data availability issue is becoming very critical for many of the blockchain, including Ethereum and Harmony ourselves. So to think about that we have done the open staking that each of the shard already be able to sync within the shard.

So in some ways, our shard already is figuring out the data availability problem in terms of doing the state sharding, network sharding and account sharding. For those that started a few years ago sharding has been the right approach, but many couldn't figure out how to do even state sharding which we started from the day one to do it and has been on mainnet for a few years.

The second big problem with all these app chains and data problem is to stay prominent. That we also is, at the state of innovation how to figure it out. It's very easy to say that you have hundred thousands of TPS, once you do some calculation the hard problem is how do you handle the data, both to sync them  and to prune them? How to compress not just for the full archival node, but for any node to be validating the transaction and for light clients on the mobile that you don't need to download any the data.

And for us we are also doing that, we are able to serve 800 million requests per day during our peak with the DeFi Kingdom. To be understanding is not just about the blockchain confirming the transition, but to be able to serve at the API level, the database level, the node syncing level are all critical problems and it started with the state, sync and stay prudent problem. And only hopefully later on next year the stateless client approach that it will be usable not just on the mobile but will be on the Chrome extension, maybe on the embedded marketplace that people can still fully confirm transactions without waiting for seconds if not download if not like centralized services like Infura. The approach is called the stateless client that we have some research there as well.

And the last point is everyone's favorite topic, I hope, zero knowledge proofs. For those that who are still interested in coming to the space that want to spend 5,10 more years. This is the only thing that we have really pushed and innovate beyond anyone that think about the long term investment, long term people coming in the space why they think that innovation is not just a quick few ideas, a few tokens and incentives.

There are cryptography in the zero knowledge proofs that some of the merkle trees and all these state syncs finally is ready for perfection. And we think the top community is actually many of the fellow students, if not many of the researchers, start from Bay area, that the group is called the 0xPARC, that we actually very closely following of their research. In particular, we think the keyless wallet even to enable many of the way of thinking you don't need to worry about your seed phrases, your passwords, how to sync between different device when you have multisig, will be very hard to figure out without advanced cryptography.

And many of them will be used that we have done research in this space. We'll release some paper here to talk about what can be, if everyone getting hacked on the central exchanges if not bridges, how can you have the self custody but without having to worry about the passwords or the seed phrases and to do social recovery?

The next thing that we still care about is the bridges technology. No matter how difficult it is, no matter how how the mistake comes up, we actually have done research and published paper that we know what will be the trade-off right now, not just having a multisig and a few backend servers  to host it, but to have on-chain security, using light clients on both of the EVM bridge chains to actually verify the transaction.

So for that we also done research paper there to talk about not just the Ethereum Bridge, but the Bitcoin bridge to do the trustless way of bridging different chains. So hopefully you guys can really look into that, become the innovations that you guys can look at.

The last part is one of the things that we are the most proud of, of the Dark Forests Games - that actually started by many of the students around here that become one of the key showcase of zero knowledge proof technology that we also fork and play with different incentives and different ways of actually incorporating into the gamification element of the incentives.

[Video]

Hopefully you guys love what we put together here. So that's our foundation. Hopefully the next half of the slides is really thinking about where are the applications?

What is the utility, what values do user really care? Hopefully a little bit more fun experiments as well. The idea is don't think about just many of these as games or incentives think of them as the economics that we can play almost as a country on thousands if not hundred thousand of users that is once in a lifetime opportunity.

Not just about issuing a currency for a new country, but how to facilitate, most of all sustain the economy for these millions of people, something called value for decades to come. So we use the word "economics" because for finance, you're really dealing with trading with the counter party, a few tens of traders, economies, really, where thousands, if not hundreds of thousands of people play a game together that believe that this is the way to do development.

The way we structure it is we know that the beginner of the Internet is domain names. And now we have a few cool things called the crypto names as well, like the .eth and what not. But the last part of it is these are all something called the radical market that I want to tell you how to tie what happened in the Web2 space, and now in the Web3 innovation together with the economic games.

The Internet domain is something that I want to announce today. We are fully developing a top level domain called .country. Something that we have purchased for sometime that we know that whether it's something called a DAO, Decentralized Autonomous Organization, or we just have a group of people doing multi-sig to hold a treasury or just a community of telegram in Discord. They actually identifed themselves as an interest group, maybe all the way to change leadership as a country.

We got the top-level domain name not just about a crypto thing, but what your browser on the mobile even IoT device will recognize. So that we are really not thinking just about how to bring DeFi to the mobile wallet.

We are really thinking about how to bring Web2 and Web3 together. And there's no way for us to force people to install pro extensions, setting up a new wallet, but to know where the Web2 is - The Web2 starting point is the domain names.

What we're doing here is we're going to really think about what are still the business? We are so worried about the few thousand users and the few millions of TVL.

No one knows the business in the domain names for the last two decades, the pickup is now is just Go Daddy - and it's a $10 billion business. We are not trying to disrupt it. We want to put GoDaddy internal domains of the Web3 stack. If you are still innovating about the blockchain games and DeFi, that is all great, because those are the primitives we need to build it with.

But the business opportunity itself is GoDaddy way of thinking about how to bring domain names to literally they have 50 million users of just creating new domain names every year. So GoDaddy business is great because not only is big but think a lot more of innovation of internet for the last two decades. The fact that they still taking their auction in escrow, much like what you would do on UniSwap for your tokens.

They really, really need to think about the domain names in the traditional business and how to deal with marketplace and auctions out there is what you have done that already on Uniswap if not OpenSea, but now for traditional domain names.

The second way to think about it is what many of the new creator is doing on something called a link tree. Again a very simple short links for any of your social media.

They don't do anything, but surely everywhere to go is what internet enables with the domain names. Linktree again has 23 million users. How many blockchain projects had more than a million users?

But for us to think a little bit broader, not just blockchain compared to the logistics and IoT, but thinking about commenting to the entire Web2. Again, not just about hoping the Web2 uses will convert to Web3 games, but physically infrastructure connecting with them is hopefully we will start with the minimum of the names.

The last one is your favorite. It's called LinkedIn, as you guys know. It is already a $23 billion business when they sold it to Microsoft back in the time.

It's even bigger now that the recurrent business is 6 billion and the advertising business is 5 billion. Why did I say is because this is the story of digital identity. They start with your names, they build links and the skillset they put on the profiles.

And the best thing the blockchain can do is make sure you commented and be able to match making. Much like Uniswap did it for the tokens. OpenSea did it for your NFT, and what a new marketplace called Radical Market can do it for humans.

Imagine a marketplace for not just interchangeable but the non-fungible humans that we can do. This may be how it could look like that could be very similar to what a link tree that many of the GenZ would actually put up.

We put out something a little bit more interesting. We call it s.country. For Stephen, I got my shortest name. It's just an s.country will launching in a few weeks and it will be fully also connected to the crypto names and we own it.

Just a number "1". Nothing else all I need the number "1". So we find ways to connect them together. We will tell the story how this becomes profile both in the Web2 space but actually it will be useful as a crypto Web3 names as well but the skillset will be more interesting.

Instead of showcasing that you are authenticated to this Twitter ID and a few followers, really tell me what is your vanity metric that you want to match with? When you go to an event like this, what do you want other people to remember you by? Is it how much fund you have raised? Definitely not just where you go to school and like what kind of technology that you know. I want to tell you, I want you to tell me how many lines of code you have written in SQL+.

And for myself OCaml. And for you, maybe how many people you've hired, how many people you've connected in this way. So this becomes the analytics of humans that will be verifiable. It's not just repetition, it's not just social connection. It's not just a long list of resume, but a few vanity metric is only the attention we have, to match, to connect, but very valuable on blockchain.

So the last one is really becoming the marketplace, whether you're tipping someone or actually using some way to do the content, you don't want recruiter, you don't want agent to help you to do any transaction.

Can you actually put in content information on the chain but actually protected by the blockchain over with some privacy if not zero knowledge proofs technology, but the fact that only those that pay and actually matching your skill, whether for love, for jobs or for opportunities, that blockchain can become the platform of change and much more.

The last one is obvious that even stating my hourly rate of $75 per hour is only the only way to future. We have no problem of one billion people on the platform we have a problem of our attention and imagination. Crypto Names is what we call it and will be the shortest, easiest way to remember it's just the number one.

The way we think about it is obviously ENS have already had done the best possible adoption. If not the most robust NFT internal  revenue and sales and trading volume but it's still about just five millions of the names.

We think it will be actually much broader for people to think that it's not about the domain names that you can ever get. We actually want it to be even auctionable. We also want you to bid on the names that you want so that all the domain names all will be gone. It will be connected to the Web2 space that you actually can identify on a social media profile.

To think about another great project other than ENS that is called Unstoppable they have figured it out. They give you the .x, .888,  aren't those very fun? Which is don't think this is end of story, even Unstoppable, again, as small as a names. Being able to raise back in time 80 million and a billion dollar valuation is what I want to give you the sense that we are just starting and the adoption is not where the trading DeFi or the games is only.

If you think about it, infrastructure, it will always come and it will always last. People actually in the domain name business, is an infrastructure builder for the Internet era to account.

The last part I still want to emphasize is we must know why soulbound NFT reputation, and in this case, what we do generalize and call it "Interoperable Memes".

This is what really Tik Tok and Reddit have done much better than blockchain. Blockchain keep talking about tokens and NFT but if you look at TikTok they don't care about the money value but how viral, how romantic they can become.

It's also what we have a few campaigns to come around the .one names as well.

That will be our last few slides. That may be how it looks like. This is our genesis blog minted on all.1.country.

And we have the radical markets, I won't go over today. Suffice to say that it's not just about how to launch another economy another product but how to sustain it. We have done well, the industry has done well the last few years. How to get a $2 trillion, $3 trillion.

What they haven't done yet is how to sustain. So they are actually much more economic theory that actually just like many of the railroad, many of the actually like country level development that they figure out that it's not about whether to give them more money, venture capital that booststrap.

They couldn't figure out how to sustain economy. I keep asking myself since the last few months why did we build such a high but also height and innovations but couldn't sustain? So I finally read a few more economic books that I skipped in my college that actually appealed theory that land tax is so important, a fair way to participate but also think about the fair price and quadratic way of doing not just voting but the taxation is going to be the key that will be innovating.

We have a few more questions here so hopefully you guys will go to this talk. Last video, I promise.

All right. Thank you, everyone. Hopefully you will go to s.1.country to check out this talk we write out a few more. Thank everyone for coming today.