Jan Xie: We need to build Web5, where Web5 = Web2 + Web3

On March 9th, at the Bitcoin Singapore 2024 conference jointly organized by the Nervos Foundation and ABCDE, Jan Xie, the Chief Architect of CKB, delivered a keynote speech titled "Bitcoin Renaissance: Why & How?".

Here is the content organized based on Jan Xie's English speech:

Why do we need a Bitcoin Renaissance?#

Why do we need a Bitcoin Renaissance? The simple answer is that everyone wants to use BTC.

Currently, there are already many Bitcoin Layer 2 projects, and users are transferring their BTC to these Layer 2 solutions and locking BTC on them. The problem is that people are also transferring BTC to other platforms, such as Ethereum, not just to orthodox Bitcoin Layer 2 solutions. Moreover, the amount of BTC on Ethereum far exceeds the amount of BTC on Bitcoin Layer 2. What's worse is that more BTC is actually stored in centralized exchanges, such as Coinbase's cold wallet, which holds about 1 million BTC.

What I want to say is that people don't just treat BTC as a store of value (SoV) tool, they also want to use BTC to make payments, participate in transactions, or do other things. Whether Bitcoin maximalists like it or not, if we do nothing, others will do it and they will offer worse solutions, because BTC flows everywhere like water.

Another demand we can see is that people want new assets. So far, the total value of assets issued on the Bitcoin blockchain is only $3 billion, far behind the total value of assets issued on other blockchains. Bitcoin is the most secure, decentralized, and censorship-resistant platform, but the total value of assets issued on this platform is the least. As a comparison, Tron's market value is about $12 billion, and the total value of assets issued on the Tron blockchain has reached $10 billion. Do you think this is a good thing or a bad thing?

What I want to emphasize again is that if we do nothing now, others will do it and they will offer worse solutions, and we can only use what they create.

After the launch of the Ethereum mainnet, Peter Todd and Greg Maxwell pointed out in 2016 that the architecture of Ethereum was problematic and not the direction this industry should follow. However, over the past 8 years, due to the lack of substantial progress in meeting real needs with Bitcoin, the entire crypto industry has turned to a direction that we do not want to see: building everything on PoS, account models are everywhere, and scaling solutions like sharding and Rollup have become mainstream. I don't think this is the path the crypto industry should take.

Bitcoin has brought many real innovations, but in recent years, the industry has abandoned these things, such as PoW, UTXO, and P2P networks.

So, why do we need a Bitcoin Renaissance now? I believe the reason is that there are indeed real needs, demands for new assets, and demands for the use cases of Bitcoin. Therefore, I hope that we can take advantage of the opportunity of the Bitcoin Renaissance to change the direction of the entire crypto industry, follow the architecture, values, and ideas of Bitcoin, and create new things in the right direction.

Some history of Bitcoin#

The values of Bitcoin, as written in the whitepaper, are a peer-to-peer electronic cash system. It is not the peer-to-contract or peer-to-sequencer systems that we often see now.

Bitcoin is based on PoW and UTXO. UTXO is a superior accounting method compared to the account model because on UTXO, we can create truly peer-to-peer assets, create anonymous assets, just like cash, rather than assets tied to contracts or arbitrary balances in accounts.

Bitcoin focuses on verification rather than computation. It emphasizes self-custody and the concept of "no key, no coin; key and coin are one."

How to build a better blockchain? The Bitcoin community has discussed many ideas. For example, Peter Todd has written a series of articles and blogs to discuss various ideas. I think three of his articles are very important:

Peter Todd's core idea is that we only need single-use seals and a blockchain composed entirely of single-use seals. We don't need anything else on the chain. We need to put all the computation and even more verification off-chain, which is what client-side validation should do. We need to do more off-chain and can build many new products off-chain, such as colored coins, RGB, Ordinal, Atomicals, and so on.

We can also create channels off-chain. Lightning Network is already a channel, but there is a lot of research on the architecture and design of channels. Interestingly, you can think of channels as client-side validation between two parties, which is also a form of client-side validation in my opinion.

The Bitcoin community also has many ideas about Layer 2 chains. There have been many discussions about Layer 2 chains, and they are not something new, and they are different from Ethereum's Layer 2, not Rollup. There are two dimensions to measure Layer 2 chains. One is the consensus mechanism, such as merged mining, staking, and sovereign chains with independent consensus mechanisms. The other dimension is how to bridge L1 and L2, called Two-Way Peg (2WP) in Bitcoin. The interesting thing is that the underlying chain and the off-chain protocols are orthogonal, such as CSV and channels. This means that we can merge them together.

The problem now is that the Bitcoin community has had many great ideas, but progress has been slow over the years. The first reason is the lack of programmability in Bitcoin, and the second reason is the characteristics of Bitcoin itself. It is very difficult to make any changes to the Bitcoin protocol. That's why we are here today.

How to achieve the Bitcoin Renaissance?#

How can we get what we want? There are three principles if we want to do anything in this space:

Meet the needs of users.
Adhere to the values of Bitcoin.
Do not rely on any soft forks/hard forks.

Fortunately, there is a demand for new applications and new assets, and there are already asset issuance protocols such as Ordinals, Runes, BRC-20, Taproot Assets, etc. We can issue assets on the Bitcoin blockchain, issue peer-to-peer, censorship-resistant assets without making any changes to Bitcoin, and without forking. The ability to issue assets on the Bitcoin blockchain is only limited by our own mindset or the ideology of Bitcoin maximalism. Issuing assets on the Bitcoin blockchain can actually bring many benefits to Bitcoin. The more assets issued on-chain, the higher the mining fee income, which means more security budget for Bitcoin in the future.

We also need a programmable layer because Bitcoin itself is limited, and we can't do many things on the Bitcoin blockchain. If we can create a layer based on Bitcoin that gives programmability to assets on the Bitcoin blockchain, we can do many things. This programmable layer will be the foundation for everything else, including scalability, privacy, and Bitcoin financial activities. This programmable layer will be the hub for assets on the Bitcoin blockchain and it is completely different from Ethereum. Ethereum has programmability, so it doesn't need a programmable layer, but Bitcoin does.

Then, we need a highway between Bitcoin and the programmable layer. We can use Two-Way Peg or use an innovative method to bridge assets between Bitcoin and the programmable layer, which I call UIB (Universal Isomorphic Binding). The idea of UIB originated from Cipher, the author of the RGB++ protocol, and universal isomorphic binding is one of the core technologies of RGB++. Cipher will explain the RGB++ protocol in detail later, so I will skip this part.

One point I want to point out here is the difference between Two-Way Peg and UIB. Two-Way Peg is when you transfer assets to an entity, and that entity sends you corresponding assets on another chain. UIB is a peer-to-peer mapping, directly establishing a relationship between UTXOs on two chains, without any intermediaries or third parties.

Once we have a programmable layer, we can build another layer on top of it to achieve scalability and privacy. We have many choices for this layer, such as client-side validation, Open Transactions, Nostr, Chaumian E-Cash, P2P markets, and so on. Then, we can connect everything with channels. We can connect L1, L2, and L3 with channels, and connect Web2 and Web3 with channels.

In this way, we have the following diagram:

Bitcoin serves as the underlying asset issuance platform, with a programmable layer on top. On this programmable layer, we build a layer for scalability and privacy. In my opinion, L3 doesn't even need to use a blockchain because a blockchain is good for transparency and global consensus, but it is harmful to scalability and privacy. This is different from the Ethereum ecosystem, where the panoramic view of the Ethereum ecosystem is adding more and more layers on top of Rollup. Why does this industry need so many blockchains? In fact, it doesn't need them. L3 should be a P2P network, and channels can connect everything.

Therefore, I believe what we want to build is Web5, which is Web2 plus Web3. Web3 tries to move everything onto the chain, but why do we need to move everything onto the chain? The internet has actually done a good job, it just has some problems. We can use P2P technology, cryptographic technology, and other new technologies to fix Web2. We don't need Web3, we don't need to put everything on the chain. With these technologies, what we want to build is actually Web5.

Summary#

Finally, let's summarize the key points of this speech:

There is indeed a demand for new applications and new assets.
If we don't build something censorship-resistant, privacy-preserving, and user-friendly, others will build something that is censorable, lacks privacy, and is not user-friendly.
Bitcoin needs a programmable Layer 2.
UIB provides assets with programmability.
Programmability is the foundation for everything else.
L3 needs to achieve scalability and privacy.
Channels connect everything.
BTC is everywhere.
Web5 > Web3