Following yesterday’s article, PolkaWorld continues to share with you the recent thoughts of Shawn, head of Parity developers.
If you want to check out the first half of the article (Web3 is more than just a ledger! Polkadot redefines the vision of the world computer!)
In today’s world of blockchain that is overwhelmed by all kinds of “L2”, “DeFi” and “AI hype”, Polkadot quietly adheres to a more ambitious goal: not to create a faster chain, nor to compete for narrative dividends with short attention spans, but to build a world computer that truly belongs to everyone.
This is not a metaphor, but a complete and serious system design. It has a multi-core decentralized CPU (PolkaVM), a censorship-resistant data lake, a verifiable storage layer (NOMT), a general operating system (CoreVM), an elastic and scalable performance architecture (Coretime), and a core protocol (JAM) that can host any application. More importantly, it is not for "building another chain", but for developers to build decentralized systems like writing ordinary applications.
This article will help you understand that Web3 is not just "old wine in a new bottle" or just the transfer of financial instruments, but a deep reshaping of computer architecture, identity control, user sovereignty and Internet infrastructure. From Web2 to Web3, from chain to application, from product to community, let's take a look at what the future world computer that Polkadot is trying to build will look like.
What does the world computer created by Polkadot look like?
What will the world computer created by Polkadot look like?
We have a decentralized CPU. This is the PolkaVM you heard about recently, which uses RISC-V. Of course, this CPU is multi-core with execution sharding technology. This is the technology we designed in Polkadot.
It could have decentralized persistent storage, like ParityDB or NOMT, a big project Rob Habermeyer is pushing to make Merkle-ized blockchain states quickly accessible.
It can also have decentralized memory, which is achieved through our distributed decentralized data lake.
It can also have the core of this computer, which is the JAM. The JAM is the core, which actually makes the computer work and run.
It will also have an operating system, which is CoreVM. CoreVM is like a Docker container, which allows you to run normal applications on this computer stack.
And of course, it will have a consensus mechanism that brings all the parts together, which are the elves. We believe that economic roll-ups, rather than ZK-based roll-ups, can actually provide scalability for this computer.
These are actually the technical specifications of the world computer. If you go to a store and want to see the world computer you're going to use, they'll try to sell you these.
These are the components of the Polkadot world computer.
Which Web2 applications can Web3 replace?
So, let’s start by looking at what the Web3 world looks like.
Is Web3 a technological evolution or a revolution? I think it is a combination of both.
I think Web3 is going to be a technical evolution. We shouldn't expect to rebuild or replace much of Web2. The Web2 internet worked well and it worked well. We should just fix the parts that didn't work. But it's also a social revolution. I think we're going to see a big shift in the way people interact with each other, with the physical world, and with the digital world, and there will be a shift in power, in social structures, in global access, and so on. So I don't think Web3 is going to be a huge change from a technical perspective, but I expect it to be a profound social change.
Web3 will not completely replace Web2. I want to emphasize again that most of the future Internet will still be built on the infrastructure of Web2. This is because Web2 still has obvious advantages in cost control, scalability, and corporate autonomy, which is reasonable and realistic.
But Web3 can indeed do better in certain specific scenarios. So the question is: what problems do Web2 companies want Web3 to help them solve?
In fact, many Web2 companies do not want to take responsibility for user identity management. They are not happy to deal with account passwords, data privacy, account loss and other issues - this is both troublesome and has legal risks. So you will find that they prefer to leave these to "single sign-on" or "federated identity system" to handle, such as logging in with Google or Microsoft account, which is why such services are popular. To put it bluntly, they just want to get the user's money, but don't want to bear the trouble of user data.
In addition, companies also want to minimize the number of intermediaries to increase profits. As we know, blockchain and DeFi can do just that, removing intermediaries in payment, subscription, second-hand transactions, etc.
Companies also hope to establish deeper interactive relationships with users. In this regard, Web3's DAO and other new social structures can provide good solutions.
At the same time, companies are also paying more and more attention to "verifiability" - they want to provide users with accurate and trustworthy promises, and the verifiable architecture of the Web3 system just meets this need.
Another example is the stability of services. Traditional companies will stop working during holidays and employees will take Christmas holidays, but the blockchain network will not "take holidays" - it runs 24 hours a day, 7 days a week. This resilience and stability is a kind of business value in itself.
At the same time, I also hope that users really "need" Web3. That is, they will actively desire to use those verifiable systems, more stable and resilient services, and privacy-protecting tools. They want to control their own data, have global and unrestricted access rights, and do not want to face the risk of content deletion and account closure at any time - these are exactly what Web3 wants to achieve.
But the reality is that I think the biggest risk to the popularity of Web3 actually comes from the promises we made. We told everyone that Web3 can achieve "self-sovereignty", "self-custody", "uncensorable services", "applications that cannot be forcibly shut down", etc. But the problem is that not everyone really wants these things.
The biggest problem is that many people actually enjoy the "convenience" of relying on centralized systems. For example, if there is a problem, you can call customer service and someone will help you solve it immediately; if you forget your password, someone can help you reset it; if you need financial services, the bank teller will help you manage your assets. Although these services mean that you hand over power to others, they do save you trouble and effort.
What we want to do in Web3 is to let everyone take control of these powers and responsibilities. But you have to ask yourself: Do people really want to be responsible for themselves?
Many people do not really realize what we call the "problem of the centralized system". It is difficult for them to understand the significance of Web3 without feeling the pain.
For example, if you talked about Web3 with your family at Christmas dinner, they might ask you in confusion: "This sounds a bit mysterious. Why should I own some kind of virtual currency?" You might explain: "This way you can have full control over your money, transfer money at any time, and don't rely on banks." But they will ask you in return: "What if I lose it?" - This question is really difficult to answer.
Because they have not experienced "systemic injustice" like us, they don't think there is anything wrong with the existing system, nor do they feel the need to replace it with an unfamiliar new system.
But if you really take the time to talk to different people, you’ll find that almost everyone has actually encountered problems at some point: loan rejection, difficulty opening a bank account, identity authentication barriers... It’s just that these "pain points" are not painful enough for them to actively look for alternatives to Web3.
We are now in a system that is firmly controlled by "centralized forces". For Web3 to really become important, it usually requires enough "collective pain" - for example, people personally experience data abuse, deprivation of rights, asset blockage, etc. But the reality is that perhaps many people will never experience the pain of "having to use Web3" in their lifetime.
Or maybe, people like me have already felt that the power of centralization has already quietly infiltrated the Web3 ecosystem we are building. If they have already controlled the world before we have time to build a truly decentralized system, then the "new world" we want may not be realized at all.
For example, people are excited to see Trump support DeFi, Goldman Sachs and JPMorgan Chase support Web3, but this is actually a dangerous illusion. Because these people and institutions are the representatives of the old world that we originally wanted to get rid of and fight against. They may repackage the old logic of centralization with a "new system that looks better" and once again take away the decentralized future that should belong to everyone. This kind of "being seized by illusion" is the most terrifying.
So I think this is one of the biggest risks facing Web3 - most people don't understand the nature of the problem and can't recognize the silent control that is happening. This is why I've put a lot of energy on "education" recently. I think education is the key to changing all this.
Through education, more people can recognize the problems in the current system. Even if they have not experienced it themselves, they can understand through learning "why we need decentralization."
For example, if I were to list the top 40 most visited websites in the world and see which of them could be replaced with Web3, I would first exclude some of the “obvious” financial scenarios, such as payment systems, subscription services, and online transactions. These Web3s already do a good job and we all know that they work.
But I want to go a step further and analyze what differences and inspirations will be brought by Web2 applications such as search engines and social networks if they are reconstructed with Web3. We have to find new innovations and landing spaces in these scenarios, rather than just staying at the repetition of the "financial level".
Let’s take a look at online search. In fact, many aspects are still the same as Web2, such as crawlers, indexing, and sorting algorithms in search engines. These links are essentially difficult to decentralize because centralized architecture is more efficient and has better results.
But now there are some changes. We are slowly transitioning from traditional search to AI-driven search experience. People are getting used to using AI to understand their questions and get more accurate answers instead of browsing a bunch of web pages.
However, in the world of Web3, searches can also become more “trustworthy.” For example:
You can have a verifiable search index to know which web pages are included in the search results and whether your content is displayed fairly;
You can also see how search results are ranked, which is not done by a company behind the scenes, but is determined by a transparent mechanism;
The sorting logic can even be managed by a DAO, for example, the DAO can select which websites are more influential and credible;
The advertising model can also be replaced by "micropayments": instead of being bombarded with ads, you can pay a few cents directly to access the content of your favorite websites.
This is actually what Brave browser has been doing. They hope that you can choose to browse the web: spend a little money and not see ads. The addition of AI has brought a new possibility: using verifiable data to train the model, users can know which of their web browsing behaviors are used to train AI search results.
That is, Web3 itself does not necessarily create completely different applications, but it can provide more transparency, verifiability, and choice.
Let’s look at applications such as video streaming (such as YouTube). Most video content does not need to be put on the blockchain, as it is too costly and inefficient. They will still be hosted on Web2 servers.
But you can imagine a streaming platform that combines Web3 features like:
You log in with your Web3 wallet;
If the video you upload is important, may be censored, or you want it to be always findable - then you can upload it to the chain or decentralized storage network;
Although this is more expensive, you gain capabilities such as censorship resistance, timestamp proof, hash verification (proving that it has not been tampered with), which are not possible on centralized platforms.
In addition, Web3 can bring more capabilities to streaming platforms:
The algorithm is open and transparent, and you know how your video is recommended;
Content filtering is user-controllable, and you can decide what you want to see and what you don’t want to see;
Users can participate in the content ecosystem through micropayments. For example, if you write subtitles or in-depth comments for a video, the platform can reward you through smart contracts.
Ultimately, the relationship between content creators and the community will be closer.
Now YouTube and Twitter are also trying to add "community interaction" features. But if we reconstruct these features with DAO and smart contracts, creators can even fully control their platforms and communities, which may give birth to a new system that is more powerful and fairer than existing platforms.
Let's talk about social media. Of course, some content, such as ordinary posts without controversy, may still be hosted on Web2 servers. But in the world of Web3, we will have some new possibilities: you are no longer a "user controlled by the platform". You can have an identity that is completely your own, that is, your account, your information, and the connection between you and your friends no longer belong to Facebook, X, Instagram or other platforms, but to you.
Your "social graph" (people you know, friends you add, and the history of your interactions) will follow you, rather than being locked in a certain platform. You can even "take" these friends to other platforms, using universal, platform-free protocols.
For example: you can "move" from Facebook to another platform, but your friends, photos, and followers can be taken with them. It's not like today where you lose everything once you cancel your account.
In addition, Web3 will also bring direct monetization capabilities to the creator economy. Creators can build their own communities and directly implement value exchange, support, rewards and other functions in them without relying on third-party platforms for commissions or audits.
At first glance, many of these features seem similar to Web2 - but the real difference is that you have full control over your identity and data, which is guaranteed by technology in Web3, not by the platform.
What are World Computer's products?
Next, let’s talk about the “product” level.
Every time I mention the idea of a “world computer”, someone asks me questions like: “Can it run Docker?” “Can it run Firefox?” “Can it run Doom or Photoshop?”
My answer is: Yes.
A real world computer should be able to run all kinds of programs that we use on everyday computers. But the question is not "can it run", but "is it necessary".
Because it costs more to use the world computer (that is, the decentralized computing platform) to run these daily programs. Are you willing to spend a lot of resources just to open a browser in a decentralized way? Most people probably won't do that.
Next, I want to talk about a misunderstanding - the current mainstream way of Web3 development, such as Substrate or Polkadot SDK, these tools have affected our understanding of how "Web3 applications" should be built.
The story we tell people is: "Want to do Web3 applications? Then build your own blockchain."
But in reality, most people don’t want or need to have their own chain. They just want to make an application that “uses Web3 technology” instead of building a chain.
The reality also proves this problem:
Many so-called Web3 applications appear to be chains, but are actually centralized;
They are often controlled by foundations or institutions (such as USDC and Tether), which can freeze accounts and prevent transactions, which runs counter to the original intention of Web3 decentralization.
Even many L2s are forced to do so: they were originally more suitable for "Web3 applications", but because the main network is too expensive and too congested, they have to build their own chain.
So we need to reflect: Is it that many projects did not originally want to build chains, but were forced to do so?
And the World Computer will change all that!
You don’t need to build a chain for your application, you just need to build the application and deploy it on the world computer. The most typical example is ledger-based apps. In the past, we would say, “If you want to do ledger, you need to build a chain.” But in the future, you can directly deploy your own ledger application on the world computer without having to worry about infrastructure.
I believe this will also be the future of Polkadot and Kusama - they will become one of the applications on the world computer, rather than the "bottom-level" system itself.
Of course, everything we do with smart contracts today, and all the projects that we are forced to build chains for today, can become ordinary applications in the future - running on the world's computer, while enjoying Web3 characteristics such as decentralization, verifiability, and censorship resistance.
Ultimately, we will re-understand Web3: not isolated chains, but a platform that allows developers to build "ordinary but decentralized applications". This is the real future.
What new application sets will JAM bring?
Next, we will talk about a new type of application brought by JAM (Join-Accumulate Machine). Their logic is very different from our traditional understanding of "blockchain applications".
Problems with traditional blockchain applications
On a chain like Ethereum, if you want to write a smart contract that does not change any on-chain state and just does addition, it actually makes no sense - you are just "burning money doing meaningless calculations", and these contracts will remain on the chain forever, wasting resources and not conducive to expansion.
What new capabilities does JAM provide?
JAM introduces the DA layer (Data Availability Layer), which you can think of as the computer's memory. This means that developers can build an application that "runs temporarily and does not write to the ledger."
That is:
You can use JAM for short-term interactions that don’t require on-chain storage.
These interactions only run in memory and disappear after execution, with only the results recorded in the ledger (if necessary).
For example, scenario 1: peer-to-peer transaction (buying a car)
In the past, we would write a smart contract on Ethereum to simulate the role of a "neutral custodian", for example: "When I receive the car keys, I will automatically transfer the money."
But once the contract is deployed, it will remain on the chain forever, even if you only trade it once. It wastes resources.
But if you use JAM’s DA layer, you can make this escrow logic only exist during execution, for example, the transaction is completed within 5 minutes, and then:
Only the final result of money circulation will be written into the chain;
The intermediary custody logic does not “live forever” or waste on-chain space.
Scenario 2: Decentralized Game Playing (Twitch Plays Pokémon)
Imagine a small game on the chain, where everyone controls a character together by sending commands (such as "up", "attack", "defense", etc.), just like "Everyone Plays Pokemon" on Switch.
JAM can handle such high-frequency, real-time interactions that do not require each step to be written to the ledger:
The vast majority of the game runs in JAM's memory;
Only key events and final results are recorded on the chain.
Now, with the capabilities provided by JAM, we don’t need to write all the data to the blockchain. For example, in a game scenario like “Everyone Plays Pokémon”: we can just let the game run in memory, and players influence the direction of the game by submitting transactions. The whole process is not written to the chain, but only in memory. When the game ends, it disappears naturally and no longer occupies resources - this will not cause any pressure on the scalability of the platform.
This is a very efficient way to keep "non-permanent information" in memory for processing.
In addition to games, there are also scenarios such as "Global Chat Room". If you just want to send a decentralized, uncensored message to the world:
There is no need to put it in the ledger;
You can use the JAM's "non-persistent state" (i.e. memory) to accomplish this;
It can achieve global communication and avoid unnecessary storage costs.
Of course, my grandest vision for the World Computer is to build a Network State. This starts with identity authentication.
Nowadays, AI is very advanced, and we often cannot tell what is real and who is a real person. In this context, we need an identity system that everyone can verify and trust.
If such systems rely on centralized institutions, they are prone to abuse. For example:
Identities can be forged;
Certain groups of people may be deprived of access.
But if we use the "world computer" on the blockchain to do this, we can build an identity system in a decentralized, verifiable, and equal way for everyone.
With “real identity”, we can build a trusted voting system — allowing everyone to participate in governance and truly influence decisions on the chain and in the real world. Once we have:
True identity
Verifiable Voting
Coordinated consensus mechanism
We can start building network nations - digital communities without geographic boundaries and based on shared values. The economic coordination and community governance of these network nations can be built on a world computer like JAM.
This brings us back to our starting point: we want to build a more resilient world. To achieve this, we need a more resilient computing platform - this is the mission of JAM and the world computer.
The Future of Polkadot
I want to end with some thoughts on the future of Polkadot.
Polkadot has always given people a feeling of being “different”. If you have been in this ecosystem for a long time, and then look at other blockchain projects on both sides, you may sigh:
“Are we really still in the same circle?”
Now, I think this "sense of difference" is becoming more and more obvious. And the reason is: Polkadot has been taking a completely different path.
If you can understand the fundamental tension between decentralized ledger systems and centralized computers, you’ll understand why Polkadot has always seemed “different.”
We are trying to build a completely new and unique computing architecture.
Although it has not yet been widely adopted and has not yet been fully proven, we believe that the world needs it.
I also think that it’s time to change the way we think about Polkadot: we should no longer think of Polkadot as just a “product”.
We used to say that Polkadot “exists to support parachains”, which is actually too narrow. This is the way of thinking in the past. Now, we need to update our thinking like upgrading technology. We should not limit Polkadot to:
Only L2 is supported
Only supports chain or ledger applications
Only serves projects that use our SDK
These boundaries are artificial limitations.
Instead, we should think of Polkadot as a community with a common goal — a community whose mission is to build the general-purpose computer of the future networked world.
This computer can run any type of application, whether it is a chain or a ledger, or even if you just want to write an ordinary app, it can also host it.
This is actually the vision that Ethereum originally wanted to achieve.
But I think Polkadot can be a community that is closer to that vision.
Maybe it’s not the most lively, or the easiest to understand, but it’s the group of people most likely to build a stronger, freer future.
Original video: https://www.youtube.com/watch?v=9tFyfYkorCw