This changes everything. More: https://twitter.com/HyperspaceOrg | Thread: https://twitter.com/varun_mathur/status/1618648357337202688
This changes everything. More: https://twitter.com/HyperspaceOrg | Thread: https://twitter.com/varun_mathur/status/1618648357337202688
Subscribe to Hyperspace
Subscribe to Hyperspace
Share Dialog
Share Dialog
<100 subscribers
<100 subscribers
I am a time-traveler from 2025. This is a moment in time when the world knows who Satoshi is and when blockchains are used on a daily basis by hundreds of millions of people. These two statements are closely related and our present day reality has its origins in the below story you are about to read..
On January 24th, 2023, a group of people collectively known simply as Gutenberg announced Hyperspace, a new type of a supercomputer powered by an entirely browser-based blockchain. In hindsight, it has gone to be one of the most consequential products in the history of computing since the January 24th, 1984 launch of the Macintosh. It sounds preposterous when you read it in 2023, but not in our present. Our world is very different than yours - and it was willed into existence by a bunch of people who had no business doing any of it.
To give you some perspective on how outrageous the Hyperspace announcement was - consider that by early 2023 there existed dozens of extremely well-funded blockchains. The community at the time felt what had to be invented was already invented, and it was a function of people not understanding the utility of blockchains. The problem was, the community was building blockchains as extensions of research projects with just enough friction points which went unnoticed, and then most others unfortunately were caught up in extracting all value they could from these systems.
We learned how to build blockchains from Satoshi, but not the why
In this world, the message of Satoshi was lost. It was a message of simplicity, of equal opportunity to all at the start, and of shared cryptoeconomic benefits across the network. It was a message of making a sacrifice for the wider good, and that sacrifice was in the form of choosing to live an ordinary life and not getting credit for your invention, and also that of deliberately not mining as much as you could have - choosing again the benefit of the network. Consider that back in 2009, a college student could download the Bitcoin client and earn 50 bitcoins for simply being a full network participant. These bitcoins were nearly worthless back then, but provided equal opportunity for all just incase the network took off.
The franchise era of blockchains which consumed the world
Back by 2022, blockchains had fully entered the franchise era, much like a local business person decides to run a McDonald’s franchise. It was one of investment: of deep financial and technical resources to be a full validator. Block production was actively moved to centralized entities by deliberate design decisions, where for Ethereum, at one point over 80% was being done by a single corporate entity. Full validators numbered in just a few thousands, which were also located mostly in specific datacenters in AWS Virginia and Germany. Many other projects masqueraded as blockchains when they really were mildly more decentralized than traditional corporate databases. A liquid staking derivative entity with around 100 members held theoretical governance power over many major Proof-of-Stake secured blockchains at the time. Transactions could be censored and a significantly worse UX provided by simply being on the list of an obscure bureaucrat of one particular country. The world had also come to associate crypto with greed, not utility, after the events of 2022.
Regular human beings were tuned out of this corporate world of blockchains except as speculators. That, was a problem.
While the world of earning full network rewards was out of reach of almost the entire world (consider in 2022 only 6% of American households could afford to be an Ethereum full validator), developers were expected to learn complex programming languages while also being able to figure out how to connect all the pieces in the blockchain ecosystem at the time. This led to the number of active blockchain developers to be in the 1000s, at a time when there existed over 100 million software developers in the world. Users were willing to make even less sacrifices due to just enough friction points. We were not building systems and apps practical enough for people in Manila or Casablanca or even Phoenix, Arizona. This led to a fundamental inequality, a world where those with sufficient technical and financial resources were the primary benefactors of the powerful new world of blockchains. It wasn’t fair.
The smart contract revolution of the 2010s was observed, but not experienced by the world at large.
Ethereum had been the most fantastic technology project of the 2010s, ushering in a wave of creativity. However the talking points of building an Uber in 2015 on Ethereum, still remained talking points even until 2022. Why wasn’t it happening ? Why did the world of high commission fees charged by proprietary database providers still used even by everyone in the blockchain community at the time ? What was widespread disruption waiting for ? Why did no other blockchain lead to practical, daily life utility for anybody ? Why were blockchains tied to the bull/bear sentiment market when the GDP wasn’t and the world population lived active economic lives ?
It turns out the need of the hour was a fresh start to re-think everything from the ground up. A new grassroots movement, a zero to one effort re-thinking everything. A passionate hobbyist team which had nothing to lose could ask ridiculous questions like:
Ok, what does a massively more decentralized blockchain than Ethereum look like ?
How do we bring back the ethos of participation amongst regular human beings like Bitcoin and BitTorrent did in the earlier years ?
And can we make the entire blockchain experience entirely browser-native for validators, users and developers ?
Why would people want to pay “franchise owners” for computation on other blockchain systems when they could simply use their own browsers ?
What system design is fair for tens of millions to use it concurrently, from Manila to Mumbai ?
Opening page of the first draft of the paper: It presents the detailed P2P browser engineering innovations, ZK-Proof and WebAssembly based runtime called the Nanochain, an uber-decentralized parallel execution and parallel block production architecture called Interstellar, with all of this leveraging a massively decentralized state subprotocol called Groot. The paper presents the mathematical and engineering basis for why the design is secure, how it works and scales, and how fast it is. It then goes beyond to explain how we massively simplify the user, developer and the validator end-to-end experience.
All zombie blockchains are fast and cheap. But our quest was for a design which was the cheapest and the fastest with peak concurrency
This is what the blockchain landscape looked like when Hyperspace was announced:
This blockchain was the “microchip“ of the overall new type of the browser-based P2P supercomputer which went on to be built in later years, and it was the fastest L1 and the most powerful design at the time of the announcement. It was a combination of fundamental new innovations and learning/iterating quickly based on industry observations which led to a parallel execution zkVM design across the network (going beyond a powerful server, the prevalent design in other blockchains) which could also do parallel validation. Set of breakthrough technologies in Hyperspace at the time included:
1. Browser P2P Networking: Enhancing browsers to run a blockchain protocol without requiring the user to download anything
The browser networking solutions the Gutenberg team figured out have been the most groundbreaking ones in a generation, and they enabled a fundamental new era of the web which is inherently peer-to-peer.
As part of the first set of engineering challenges, the team developed a networking library which enhanced existing browsers to provide full peer-to-peer connectivity between browser nodes: from discovery of such nodes to forming stable connections to exchange data. To run a full blockchain node now it no longer required to download and install software on your computer. This now made it inevitable that future blockchains and decentralized services would be fully browser-based in the near future and it would open up a world of decentralized web apps. Imagine entire front-end of such apps running on this P2P network, with no need of big tech cloud services. This enabled better privacy for users, lower latency in some scenarios, and eventually lower costs along with providing a perpetual running, highly available permissionless network.
PS: If you liked Nostr in late 2022, you would have loved the Hyperspace networking protocol in 2023 as it provided similar benefits but now with full P2P.
2. Web zkVM: Building a blazing fast secure runtime in the browser which could generate a Proof of Execution based on ZK + WebAssembly
Hyperspace utilizes the user’s own browser in a new web zkVM to execute smart contracts and capture the Proof of Execution Trace in a STARK. These STARKs (“actions”) are bundled together into a single SNARK transaction which is propagated through the network. This design enables massively parallel execution. Since it is already known which specific states are impacted, this then plays a pivotal role in parallel validation. Overall, this design enables 10ms-50ms pre-L1-finality in the same region, providing an experience even faster than what was known as web2 and enabling new type of web3 multiplayer use cases.
Here’s a link to a real demo of the Hyperspace web zkVM in action - try it out!
This website has an embedded Zero Knowledge runtime. To see just how fast it is, try and test the limits on our simulation. When you hit the play button, during the next ten seconds, send as many actions to it as you can by pressing any key/any mouse click or tap. Every time you trigger an action the runtime will execute a piece of WebAssembly (WASM) code that emulates fetching/modifying state and then generates it’s Proof of Execution (making the square green). Then all these proofs are merged into a single proof, which is what constitutes a transaction in the Hyperspace system.
If you find this fast enough - imagine this as the foundation for the UX for social, gaming, financial and marketplace apps.
3. Parallel validation: A multi-lane highway approach to block building and validation which was pragmatic for global usage
Narwhal-inspired DAG and Enhanced Hotstuff A set of decentralized nodes utilizing a DAG and a weighted algorithm efficiently package transactions into blocks which are sent to specific validator shards.
Conflict-free sharded consensus These shards could also receive transactions directly, building further on the blocks, and perform decentralized consensus leading to multiple micro-blocks validated in parallel, without conflict.
Breakthrough design for sharded state Hyperspace was also designed from the outset with a sharded state strategy which would prevent the state bloat problem at-scale which impacted all monolithic L1 constructions till date. Ethereum’s own tech project lead had termed it as being in a “death spiral“ due to state growing by 1 GB/day. Hyperspace was designed to solve this problem from the outset.
Optimized for cross-program invocations Overall the system was designed for the worst case: maximum cross-program invocations. This required re-imagining the whole validation and execution layer, by completely removing the execution bottleneck from validators which in turn enabled parallel validation achieving true platform scalability.
Inherent horizontal scaling design The system design enabled continuous addition of validator shards as the system grew, with the TPS increasing with it since it did not require high system requirements from an individual block producing node. This had been a foundational element of the Gutenberg team’s philosophy: enabling regular people with simple laptops and just browsers to be full network participants.
4. Hyperspace enshrined everything from identity, oracle and even re-staking into the base protocol enabling premium UX and DX
There is something common between the approach Apple takes with its products and what Vitalik suggested how a new blockchain should be built: integrated experience.
Enshrined oracleHyperspace enshrined an oracle into the core protocol which the existing set of validators could join using the browser-based client software to start providing oracle services to the network. By utilizing re-staking mechanism enshrined into the base protocol, it improved security and range of services available such as crosschain primitives, automation, timekeeping and web2 data. This construction also made it extremely easy for dApps that were built on Ethereum, Solana, Polkadot and Near to directly integrate with Hyperspace. From a developer perspective, this enshrined oracle provided low-level primitives that simplified working with off-chain data compared to the oracles which existed back then.
Enshrined identity, smart wallet and NFT standardsHyperspace provided an integrated identity and smart wallet subprotocol which introduced a new way of managing separate identities across the Hyperspace ecosystem. These are called session subwallets which derive cryptographically from a user’s main wallet and abstracts them to multiple identities for their specific apps. This design was inspired from traditional web UX, and enabled users to stay in full control of what application they give permissions and spending limits to. Overall this zero install approach eased adoption for beginners while providing pragmatic escalated security options to power users. This subprotocol also enshrined token/NFT standards, making it simpler for smart contract developers to integrate into it while enabling composability. This unlocked new design spaces of confirm-free transactions that were perfectly tailored for multiplayer scenarios.
5. Designed for developers. “The line of code that is the fastest to write, that never breaks, that never needs maintenance is the line that you never have to write.” - Steve Jobs, at a 1997 keynote.
Hyperspace focused on TypeScript as the primary smart contract development language, which combined with the SDK, enabled millions of hobbyist web developers to get comfortable with smart contracts very quickly. The protocol also enshrined multiple subprotocols and decentralized services to provide an integrated experience to developers which simplified building complex applications.
The design of a smart contract wallet which provided context-driven access turned out to be beneficial for developers as it provided security to users, building trust and enabling more usage. Overall, Hyperspace had been designed for the fastest blockchain experience, where the pre-L1-finality was comparable to the speed of a Google Search in the same region.
6. As Uber disrupted the taxi medallion business, Hyperspace disrupted the full node validator business - democratizing it
Hyperspace cryptoeconomics was driven by fundamental ideologies, which included:
Even a college student with a laptop should be able to earn significant rewards by simply using the browser and participating fully in this new network.
Deflationary model and which with enshrined re-staking accrues values to the base token, thereby increasing the security of the network.
Keeping the blockchain fiercely independent from any governance mechanism imposed by an external liquid staking derivative. If all other PoS blockchains are impacted by censorship for any reason, at least Hyperspace should remain viable.
Maximally decentralizing block production amongst regular human beings, and not corporate entities which could censor and degrade the permissionless UX.
The idea of a “supercomputer” used to be this extremely powerful set of computers connected to each other physically and all located in one building. Hyperspace takes that concept and applies it to the level of hundreds of millions of people simply using browsers on their laptops, with zero installation required across the board. That fully P2P “computer“ builds on three inevitable trends of the 2020s:
Laptops and other personal computing devices getting insanely powerful and cheaper at the same time. It makes running computations in datacenters obsolete.
Bandwidth getting insanely faster and cheaper at the same time, and with improved browser P2P networking unlocking new capabilities.
Era of Zero Knowledge Proofs and WebAssembly, which enables near-native speed on a computer even from within the browser, with no installation required.
Hyperspace is how regular human beings experience the power of a massively efficient network of browser nodes connected to each other directly and be able to do a wide variety of functions typically done using proprietary API and cloud providers. The “Shopify, Uber, Airbnb and even ChatGPT” of our era increasingly and inevitably run on such a P2P network maintained by regular web users, and not be beholden to one-specific-company and its rules. Software functions have gotten massively commoditized, and it is the age of the smart contract.
Here’s the world we imagine: everything has crypto
With Hyperspace you now don’t need to pay a third-party closed database a high take rate just for execution of business logic. Code is open, alive and available. You can efficiently transact with other people directly, using your own computer. You can seamlessly tap into millions of smart contracts, or compose your own quickly. Your data is not locked in; you own and control it. You can also choose to participate in this network and provide your computer’s resources to it, improving the health of the network and earning rewards. Think BitTorrent.
This is the world of peer-to-peer computing powered by the browser, requiring zero downloads. The Hyperspace blockchain is like the microchip, running invisibly - as a user you experience the power of the full network.
We are making the fundamental web experience better.
"Web is the platform" - Marc Andreessen
The mainstream browser vendors have not been fully aligned with a peer-to-peer computing paradigm, even as they continued making the underlying hardware and software exceptionally powerful. This required the Gutenberg team to come up with multiple browser-networking related innovations which involved some trade-offs. Ultimately, in the quest to provide the most premium and native web3 experience feasible to users, the ultimate goal has been to build out the Hyperbrowser.
As a dedicated browser built on top of Chromium, it would enable bringing the zk-proving runtime to near-native performance, and even enabling the home stakers to earn more by providing access to the GPU and storage directly. This browser will go beyond limitations in Chromium's networking stack to be natively peer-to-peer, and make authentication with the enshrined identity layer even more convenient for the end users. Imagine being able to directly type someone’s hydentity address in the browser and being able to view their profile, send tokens, NFTs and more. The .hy primitive would be natively integrated into the Hyperbrowser.
Decentralized Web Apps would load much faster due to the runtime being native in the browser as well. This works by telling the server where the dApp is stored and that the requester is using the Hyperbrowser. The server then ships a smaller app that only includes the smart contract, CSS, HTML and Javascript - with no extra runtime. This would save on bandwidth by massively reducing the size needed to load the app and would provide an even faster experience while improving decentralization. A world of tens of millions of smart contracts being native web objects is fundamentally good for consumers.
Overall, Hyperspace intends to raise the bar for what it means to be a web browser in the 2020s. The client-server browsers of the 2010s are great, but it is time now for insanely powerful peer-to-peer software across devices.
A quick introduction to how the world of software works now
Now in 2025, the creation of software is done by the several generation of AI products which came after ChatGPT and similar models, and the deployment and usage of software happens on a turbocharged P2P “supercomputer“ like Hyperspace. Your AI agent crafts end-to-end personalized software experiences for you based on your history and basic prompts and uses Hyperspace as the deploy layer.
Here is an example of a user journey of renting out their spare room:
In order to simply rent out a room to travelers, you don’t have to go on a website, click many buttons, manually choose prices, have it charge a significant and arbitrary commission fees, while also locking your data in! Instead, here is what happens: your AI agent already knows your intent and schedule, finds the most reputable and cost-effective auction smart contracts amongst the millions on Hyperspace, and packages it in multiple types of fully functional software experiences for you, using your data made available by you - with all of this entirely running on the fully peer-to-peer network of simply tens of millions of browser-based nodes stitched together using cryptoeconomic incentives by Hyperspace.
Even these millions of smart contracts were not built by blockchain rocket scientists as used to be the case in the early 2020s, but by AI models using simple prompts from regular web users or by tens of millions of TypeScript developers. By saying a single sentence, you can now rent out your room for the best price and with near 0% commission fees as the smart contracts auto-compete to provide the service to you! We went from advertisers auto-competing in Google search results to smart contracts auto-competing on Hyperspace to provide you the best software utility.
Software became commoditized like electricity. Other use cases included:
P2P Discord and other messaging apps.
P2P Netflix, Youtube and other filesharing systems.
P2P Notion and other note-taking systems.
P2P Metaverse, enabled further by tech such as WebGPU and WebXR. MMORPGs where everyone is moving around through a map to a 3D world of connected apps.
P2P Gaming: where one player hosts and others can join directly.
P2P AI: Each user contributing to training a model using just their local user data.
The idea of a single company owning your data now seems like a distant memory. We have extremely powerful laptops, which are seamlessly connected to each other using the Hyperspace Supercomputer through the blazing fast and cheap bandwidth now available. This is our new world, and it’s exciting as we are still in the early days of realizing the full power of such a network.
Gutenberg, the inventor of the printing press, was a crazy driven fellow. The single-minded pursuit of bringing that invention to market 500 years ago changed the course of humanity for the better. We learnt to pass on knowledge to each other using books beyond our lifetimes, and that tapped into the creative potential of the collective human mind. We have kept on getting better.
Blockchains are the next order of magnitude invention of our times. We have only scratched the surface, but imagine what happens when we use them at-scale, across hundreds of millions of us at least, as our collective truth machine across time.
Our civilization advances every time we use a common language
Think about the 26 letters of the English alphabet - how powerful has that been for constructing ever more complex and useful higher dimension experiences ? Now think about the billions of such alphabets added by billions of people in a common agreed framework, and we realize that we still are in early days of realizing the power of compounding experiences here. Just like we don’t log in individually to each file server to retrieve a file like we did until the early ‘90s, preferring the dramatic ease and convenience of the interconnected web, that is the end-state of this new hyper-connected world as well. We, humans and AI bots, all speak the same language, and our vocabulary just expanded exponentially.
The pursuit of Hyperspace was not academic. The Gutenberg team did not care about technique and industry as much as it cared about bottomline utility. The intent here was to build out the printing press of the next several decades, which runs on the computing devices owned by people directly. The world needed this.
AI models being able to choose the winning smart contract amongst millions of options turned out to be a much better model for consumers than similar technology functions being locked in and made available by a few startups only. This fundamental flippening of the marketplace model has now led to software functions being priced like electricity. This has saved people tens of billions of dollars so far, improved privacy, and increased choice. Useful software is now available just like we moved on from renting each movie individually to paying a simplified fees for near unlimited titles on Netflix.
Blockchains are efficient market platforms where competition leads to the best outcome for consumers
We are literal nobodies. If you look into our backgrounds, you will not believe we could have architected the fastest and the most decentralized smart contract blockchain till date. Instead, we were simply insanely driven - and we learnt and applied what we could. We were coming from a perspective of sheer frustration from a validator, user and developer perspective. Observing and learning about Satoshi provided just enough motivation and guidance as well that anybody in the world can self-study and train and aspire to take on the most challenging projects there can be. Our challenge was building a massively more decentralized system which was fair and fast for people around the world even when tens of millions use it at the same time. And our work is not done - we are on a quest to build out the most remarkable computing platform of this decade which fundamentally improves human life.
We are mostly driven by the need to do this. In the pursuit of doing this the right way over an extended period of time, we chose to take an earnest path to fundraising. This resulted in us ending up with a very distributed set of investors, with nobody holding any governance rights on how this blockchain gets built out. We are excited that we have a shot at being the first new, “by the people, for the people“ blockchain of this era, and then going beyond and realizing the supercomputer vision where millions of browser nodes are seamlessly stitched together using cryptoeconomic incentives.
Check out our website for a quick overview of the core L1 blockchain.
You can earn a fun badge for being an early supporter of Hyperspace by simply claiming your identity handle over at this link (hy.gl) which you can use later on when the full network launches.
If you want a sneak peek at the detailed technical documentation, please DM.
Contribute to protocol research - stay tuned for announcements related to this.
Contests: hop over to our Discord to stay updated with these upcoming contests: Billboards: submit clever and simple billboard ideas like you see in this post - we will be selecting multiple winners frequently and giving away real-world prizes. Name the L1 blockchain: We would be opening up the naming of the core L1 blockchain to a public opinion poll.Build stuff using our SDK Alpha Release: Coming up soon enough we will be demo-ing a game built using our SDK and making it available for tinkering around.
In the days ahead - you will hear from the Gutenberg team a lot: we have a lot of demos, talks, fun guides, essays and our detailed paper coming up for release presenting a mathematical, engineering and conceptual basis for our system and cryptoeconomics design. We have an open canvas, a lot to do still, and would love to hear from you..
I am a time-traveler from 2025. This is a moment in time when the world knows who Satoshi is and when blockchains are used on a daily basis by hundreds of millions of people. These two statements are closely related and our present day reality has its origins in the below story you are about to read..
On January 24th, 2023, a group of people collectively known simply as Gutenberg announced Hyperspace, a new type of a supercomputer powered by an entirely browser-based blockchain. In hindsight, it has gone to be one of the most consequential products in the history of computing since the January 24th, 1984 launch of the Macintosh. It sounds preposterous when you read it in 2023, but not in our present. Our world is very different than yours - and it was willed into existence by a bunch of people who had no business doing any of it.
To give you some perspective on how outrageous the Hyperspace announcement was - consider that by early 2023 there existed dozens of extremely well-funded blockchains. The community at the time felt what had to be invented was already invented, and it was a function of people not understanding the utility of blockchains. The problem was, the community was building blockchains as extensions of research projects with just enough friction points which went unnoticed, and then most others unfortunately were caught up in extracting all value they could from these systems.
We learned how to build blockchains from Satoshi, but not the why
In this world, the message of Satoshi was lost. It was a message of simplicity, of equal opportunity to all at the start, and of shared cryptoeconomic benefits across the network. It was a message of making a sacrifice for the wider good, and that sacrifice was in the form of choosing to live an ordinary life and not getting credit for your invention, and also that of deliberately not mining as much as you could have - choosing again the benefit of the network. Consider that back in 2009, a college student could download the Bitcoin client and earn 50 bitcoins for simply being a full network participant. These bitcoins were nearly worthless back then, but provided equal opportunity for all just incase the network took off.
The franchise era of blockchains which consumed the world
Back by 2022, blockchains had fully entered the franchise era, much like a local business person decides to run a McDonald’s franchise. It was one of investment: of deep financial and technical resources to be a full validator. Block production was actively moved to centralized entities by deliberate design decisions, where for Ethereum, at one point over 80% was being done by a single corporate entity. Full validators numbered in just a few thousands, which were also located mostly in specific datacenters in AWS Virginia and Germany. Many other projects masqueraded as blockchains when they really were mildly more decentralized than traditional corporate databases. A liquid staking derivative entity with around 100 members held theoretical governance power over many major Proof-of-Stake secured blockchains at the time. Transactions could be censored and a significantly worse UX provided by simply being on the list of an obscure bureaucrat of one particular country. The world had also come to associate crypto with greed, not utility, after the events of 2022.
Regular human beings were tuned out of this corporate world of blockchains except as speculators. That, was a problem.
While the world of earning full network rewards was out of reach of almost the entire world (consider in 2022 only 6% of American households could afford to be an Ethereum full validator), developers were expected to learn complex programming languages while also being able to figure out how to connect all the pieces in the blockchain ecosystem at the time. This led to the number of active blockchain developers to be in the 1000s, at a time when there existed over 100 million software developers in the world. Users were willing to make even less sacrifices due to just enough friction points. We were not building systems and apps practical enough for people in Manila or Casablanca or even Phoenix, Arizona. This led to a fundamental inequality, a world where those with sufficient technical and financial resources were the primary benefactors of the powerful new world of blockchains. It wasn’t fair.
The smart contract revolution of the 2010s was observed, but not experienced by the world at large.
Ethereum had been the most fantastic technology project of the 2010s, ushering in a wave of creativity. However the talking points of building an Uber in 2015 on Ethereum, still remained talking points even until 2022. Why wasn’t it happening ? Why did the world of high commission fees charged by proprietary database providers still used even by everyone in the blockchain community at the time ? What was widespread disruption waiting for ? Why did no other blockchain lead to practical, daily life utility for anybody ? Why were blockchains tied to the bull/bear sentiment market when the GDP wasn’t and the world population lived active economic lives ?
It turns out the need of the hour was a fresh start to re-think everything from the ground up. A new grassroots movement, a zero to one effort re-thinking everything. A passionate hobbyist team which had nothing to lose could ask ridiculous questions like:
Ok, what does a massively more decentralized blockchain than Ethereum look like ?
How do we bring back the ethos of participation amongst regular human beings like Bitcoin and BitTorrent did in the earlier years ?
And can we make the entire blockchain experience entirely browser-native for validators, users and developers ?
Why would people want to pay “franchise owners” for computation on other blockchain systems when they could simply use their own browsers ?
What system design is fair for tens of millions to use it concurrently, from Manila to Mumbai ?
Opening page of the first draft of the paper: It presents the detailed P2P browser engineering innovations, ZK-Proof and WebAssembly based runtime called the Nanochain, an uber-decentralized parallel execution and parallel block production architecture called Interstellar, with all of this leveraging a massively decentralized state subprotocol called Groot. The paper presents the mathematical and engineering basis for why the design is secure, how it works and scales, and how fast it is. It then goes beyond to explain how we massively simplify the user, developer and the validator end-to-end experience.
All zombie blockchains are fast and cheap. But our quest was for a design which was the cheapest and the fastest with peak concurrency
This is what the blockchain landscape looked like when Hyperspace was announced:
This blockchain was the “microchip“ of the overall new type of the browser-based P2P supercomputer which went on to be built in later years, and it was the fastest L1 and the most powerful design at the time of the announcement. It was a combination of fundamental new innovations and learning/iterating quickly based on industry observations which led to a parallel execution zkVM design across the network (going beyond a powerful server, the prevalent design in other blockchains) which could also do parallel validation. Set of breakthrough technologies in Hyperspace at the time included:
1. Browser P2P Networking: Enhancing browsers to run a blockchain protocol without requiring the user to download anything
The browser networking solutions the Gutenberg team figured out have been the most groundbreaking ones in a generation, and they enabled a fundamental new era of the web which is inherently peer-to-peer.
As part of the first set of engineering challenges, the team developed a networking library which enhanced existing browsers to provide full peer-to-peer connectivity between browser nodes: from discovery of such nodes to forming stable connections to exchange data. To run a full blockchain node now it no longer required to download and install software on your computer. This now made it inevitable that future blockchains and decentralized services would be fully browser-based in the near future and it would open up a world of decentralized web apps. Imagine entire front-end of such apps running on this P2P network, with no need of big tech cloud services. This enabled better privacy for users, lower latency in some scenarios, and eventually lower costs along with providing a perpetual running, highly available permissionless network.
PS: If you liked Nostr in late 2022, you would have loved the Hyperspace networking protocol in 2023 as it provided similar benefits but now with full P2P.
2. Web zkVM: Building a blazing fast secure runtime in the browser which could generate a Proof of Execution based on ZK + WebAssembly
Hyperspace utilizes the user’s own browser in a new web zkVM to execute smart contracts and capture the Proof of Execution Trace in a STARK. These STARKs (“actions”) are bundled together into a single SNARK transaction which is propagated through the network. This design enables massively parallel execution. Since it is already known which specific states are impacted, this then plays a pivotal role in parallel validation. Overall, this design enables 10ms-50ms pre-L1-finality in the same region, providing an experience even faster than what was known as web2 and enabling new type of web3 multiplayer use cases.
Here’s a link to a real demo of the Hyperspace web zkVM in action - try it out!
This website has an embedded Zero Knowledge runtime. To see just how fast it is, try and test the limits on our simulation. When you hit the play button, during the next ten seconds, send as many actions to it as you can by pressing any key/any mouse click or tap. Every time you trigger an action the runtime will execute a piece of WebAssembly (WASM) code that emulates fetching/modifying state and then generates it’s Proof of Execution (making the square green). Then all these proofs are merged into a single proof, which is what constitutes a transaction in the Hyperspace system.
If you find this fast enough - imagine this as the foundation for the UX for social, gaming, financial and marketplace apps.
3. Parallel validation: A multi-lane highway approach to block building and validation which was pragmatic for global usage
Narwhal-inspired DAG and Enhanced Hotstuff A set of decentralized nodes utilizing a DAG and a weighted algorithm efficiently package transactions into blocks which are sent to specific validator shards.
Conflict-free sharded consensus These shards could also receive transactions directly, building further on the blocks, and perform decentralized consensus leading to multiple micro-blocks validated in parallel, without conflict.
Breakthrough design for sharded state Hyperspace was also designed from the outset with a sharded state strategy which would prevent the state bloat problem at-scale which impacted all monolithic L1 constructions till date. Ethereum’s own tech project lead had termed it as being in a “death spiral“ due to state growing by 1 GB/day. Hyperspace was designed to solve this problem from the outset.
Optimized for cross-program invocations Overall the system was designed for the worst case: maximum cross-program invocations. This required re-imagining the whole validation and execution layer, by completely removing the execution bottleneck from validators which in turn enabled parallel validation achieving true platform scalability.
Inherent horizontal scaling design The system design enabled continuous addition of validator shards as the system grew, with the TPS increasing with it since it did not require high system requirements from an individual block producing node. This had been a foundational element of the Gutenberg team’s philosophy: enabling regular people with simple laptops and just browsers to be full network participants.
4. Hyperspace enshrined everything from identity, oracle and even re-staking into the base protocol enabling premium UX and DX
There is something common between the approach Apple takes with its products and what Vitalik suggested how a new blockchain should be built: integrated experience.
Enshrined oracleHyperspace enshrined an oracle into the core protocol which the existing set of validators could join using the browser-based client software to start providing oracle services to the network. By utilizing re-staking mechanism enshrined into the base protocol, it improved security and range of services available such as crosschain primitives, automation, timekeeping and web2 data. This construction also made it extremely easy for dApps that were built on Ethereum, Solana, Polkadot and Near to directly integrate with Hyperspace. From a developer perspective, this enshrined oracle provided low-level primitives that simplified working with off-chain data compared to the oracles which existed back then.
Enshrined identity, smart wallet and NFT standardsHyperspace provided an integrated identity and smart wallet subprotocol which introduced a new way of managing separate identities across the Hyperspace ecosystem. These are called session subwallets which derive cryptographically from a user’s main wallet and abstracts them to multiple identities for their specific apps. This design was inspired from traditional web UX, and enabled users to stay in full control of what application they give permissions and spending limits to. Overall this zero install approach eased adoption for beginners while providing pragmatic escalated security options to power users. This subprotocol also enshrined token/NFT standards, making it simpler for smart contract developers to integrate into it while enabling composability. This unlocked new design spaces of confirm-free transactions that were perfectly tailored for multiplayer scenarios.
5. Designed for developers. “The line of code that is the fastest to write, that never breaks, that never needs maintenance is the line that you never have to write.” - Steve Jobs, at a 1997 keynote.
Hyperspace focused on TypeScript as the primary smart contract development language, which combined with the SDK, enabled millions of hobbyist web developers to get comfortable with smart contracts very quickly. The protocol also enshrined multiple subprotocols and decentralized services to provide an integrated experience to developers which simplified building complex applications.
The design of a smart contract wallet which provided context-driven access turned out to be beneficial for developers as it provided security to users, building trust and enabling more usage. Overall, Hyperspace had been designed for the fastest blockchain experience, where the pre-L1-finality was comparable to the speed of a Google Search in the same region.
6. As Uber disrupted the taxi medallion business, Hyperspace disrupted the full node validator business - democratizing it
Hyperspace cryptoeconomics was driven by fundamental ideologies, which included:
Even a college student with a laptop should be able to earn significant rewards by simply using the browser and participating fully in this new network.
Deflationary model and which with enshrined re-staking accrues values to the base token, thereby increasing the security of the network.
Keeping the blockchain fiercely independent from any governance mechanism imposed by an external liquid staking derivative. If all other PoS blockchains are impacted by censorship for any reason, at least Hyperspace should remain viable.
Maximally decentralizing block production amongst regular human beings, and not corporate entities which could censor and degrade the permissionless UX.
The idea of a “supercomputer” used to be this extremely powerful set of computers connected to each other physically and all located in one building. Hyperspace takes that concept and applies it to the level of hundreds of millions of people simply using browsers on their laptops, with zero installation required across the board. That fully P2P “computer“ builds on three inevitable trends of the 2020s:
Laptops and other personal computing devices getting insanely powerful and cheaper at the same time. It makes running computations in datacenters obsolete.
Bandwidth getting insanely faster and cheaper at the same time, and with improved browser P2P networking unlocking new capabilities.
Era of Zero Knowledge Proofs and WebAssembly, which enables near-native speed on a computer even from within the browser, with no installation required.
Hyperspace is how regular human beings experience the power of a massively efficient network of browser nodes connected to each other directly and be able to do a wide variety of functions typically done using proprietary API and cloud providers. The “Shopify, Uber, Airbnb and even ChatGPT” of our era increasingly and inevitably run on such a P2P network maintained by regular web users, and not be beholden to one-specific-company and its rules. Software functions have gotten massively commoditized, and it is the age of the smart contract.
Here’s the world we imagine: everything has crypto
With Hyperspace you now don’t need to pay a third-party closed database a high take rate just for execution of business logic. Code is open, alive and available. You can efficiently transact with other people directly, using your own computer. You can seamlessly tap into millions of smart contracts, or compose your own quickly. Your data is not locked in; you own and control it. You can also choose to participate in this network and provide your computer’s resources to it, improving the health of the network and earning rewards. Think BitTorrent.
This is the world of peer-to-peer computing powered by the browser, requiring zero downloads. The Hyperspace blockchain is like the microchip, running invisibly - as a user you experience the power of the full network.
We are making the fundamental web experience better.
"Web is the platform" - Marc Andreessen
The mainstream browser vendors have not been fully aligned with a peer-to-peer computing paradigm, even as they continued making the underlying hardware and software exceptionally powerful. This required the Gutenberg team to come up with multiple browser-networking related innovations which involved some trade-offs. Ultimately, in the quest to provide the most premium and native web3 experience feasible to users, the ultimate goal has been to build out the Hyperbrowser.
As a dedicated browser built on top of Chromium, it would enable bringing the zk-proving runtime to near-native performance, and even enabling the home stakers to earn more by providing access to the GPU and storage directly. This browser will go beyond limitations in Chromium's networking stack to be natively peer-to-peer, and make authentication with the enshrined identity layer even more convenient for the end users. Imagine being able to directly type someone’s hydentity address in the browser and being able to view their profile, send tokens, NFTs and more. The .hy primitive would be natively integrated into the Hyperbrowser.
Decentralized Web Apps would load much faster due to the runtime being native in the browser as well. This works by telling the server where the dApp is stored and that the requester is using the Hyperbrowser. The server then ships a smaller app that only includes the smart contract, CSS, HTML and Javascript - with no extra runtime. This would save on bandwidth by massively reducing the size needed to load the app and would provide an even faster experience while improving decentralization. A world of tens of millions of smart contracts being native web objects is fundamentally good for consumers.
Overall, Hyperspace intends to raise the bar for what it means to be a web browser in the 2020s. The client-server browsers of the 2010s are great, but it is time now for insanely powerful peer-to-peer software across devices.
A quick introduction to how the world of software works now
Now in 2025, the creation of software is done by the several generation of AI products which came after ChatGPT and similar models, and the deployment and usage of software happens on a turbocharged P2P “supercomputer“ like Hyperspace. Your AI agent crafts end-to-end personalized software experiences for you based on your history and basic prompts and uses Hyperspace as the deploy layer.
Here is an example of a user journey of renting out their spare room:
In order to simply rent out a room to travelers, you don’t have to go on a website, click many buttons, manually choose prices, have it charge a significant and arbitrary commission fees, while also locking your data in! Instead, here is what happens: your AI agent already knows your intent and schedule, finds the most reputable and cost-effective auction smart contracts amongst the millions on Hyperspace, and packages it in multiple types of fully functional software experiences for you, using your data made available by you - with all of this entirely running on the fully peer-to-peer network of simply tens of millions of browser-based nodes stitched together using cryptoeconomic incentives by Hyperspace.
Even these millions of smart contracts were not built by blockchain rocket scientists as used to be the case in the early 2020s, but by AI models using simple prompts from regular web users or by tens of millions of TypeScript developers. By saying a single sentence, you can now rent out your room for the best price and with near 0% commission fees as the smart contracts auto-compete to provide the service to you! We went from advertisers auto-competing in Google search results to smart contracts auto-competing on Hyperspace to provide you the best software utility.
Software became commoditized like electricity. Other use cases included:
P2P Discord and other messaging apps.
P2P Netflix, Youtube and other filesharing systems.
P2P Notion and other note-taking systems.
P2P Metaverse, enabled further by tech such as WebGPU and WebXR. MMORPGs where everyone is moving around through a map to a 3D world of connected apps.
P2P Gaming: where one player hosts and others can join directly.
P2P AI: Each user contributing to training a model using just their local user data.
The idea of a single company owning your data now seems like a distant memory. We have extremely powerful laptops, which are seamlessly connected to each other using the Hyperspace Supercomputer through the blazing fast and cheap bandwidth now available. This is our new world, and it’s exciting as we are still in the early days of realizing the full power of such a network.
Gutenberg, the inventor of the printing press, was a crazy driven fellow. The single-minded pursuit of bringing that invention to market 500 years ago changed the course of humanity for the better. We learnt to pass on knowledge to each other using books beyond our lifetimes, and that tapped into the creative potential of the collective human mind. We have kept on getting better.
Blockchains are the next order of magnitude invention of our times. We have only scratched the surface, but imagine what happens when we use them at-scale, across hundreds of millions of us at least, as our collective truth machine across time.
Our civilization advances every time we use a common language
Think about the 26 letters of the English alphabet - how powerful has that been for constructing ever more complex and useful higher dimension experiences ? Now think about the billions of such alphabets added by billions of people in a common agreed framework, and we realize that we still are in early days of realizing the power of compounding experiences here. Just like we don’t log in individually to each file server to retrieve a file like we did until the early ‘90s, preferring the dramatic ease and convenience of the interconnected web, that is the end-state of this new hyper-connected world as well. We, humans and AI bots, all speak the same language, and our vocabulary just expanded exponentially.
The pursuit of Hyperspace was not academic. The Gutenberg team did not care about technique and industry as much as it cared about bottomline utility. The intent here was to build out the printing press of the next several decades, which runs on the computing devices owned by people directly. The world needed this.
AI models being able to choose the winning smart contract amongst millions of options turned out to be a much better model for consumers than similar technology functions being locked in and made available by a few startups only. This fundamental flippening of the marketplace model has now led to software functions being priced like electricity. This has saved people tens of billions of dollars so far, improved privacy, and increased choice. Useful software is now available just like we moved on from renting each movie individually to paying a simplified fees for near unlimited titles on Netflix.
Blockchains are efficient market platforms where competition leads to the best outcome for consumers
We are literal nobodies. If you look into our backgrounds, you will not believe we could have architected the fastest and the most decentralized smart contract blockchain till date. Instead, we were simply insanely driven - and we learnt and applied what we could. We were coming from a perspective of sheer frustration from a validator, user and developer perspective. Observing and learning about Satoshi provided just enough motivation and guidance as well that anybody in the world can self-study and train and aspire to take on the most challenging projects there can be. Our challenge was building a massively more decentralized system which was fair and fast for people around the world even when tens of millions use it at the same time. And our work is not done - we are on a quest to build out the most remarkable computing platform of this decade which fundamentally improves human life.
We are mostly driven by the need to do this. In the pursuit of doing this the right way over an extended period of time, we chose to take an earnest path to fundraising. This resulted in us ending up with a very distributed set of investors, with nobody holding any governance rights on how this blockchain gets built out. We are excited that we have a shot at being the first new, “by the people, for the people“ blockchain of this era, and then going beyond and realizing the supercomputer vision where millions of browser nodes are seamlessly stitched together using cryptoeconomic incentives.
Check out our website for a quick overview of the core L1 blockchain.
You can earn a fun badge for being an early supporter of Hyperspace by simply claiming your identity handle over at this link (hy.gl) which you can use later on when the full network launches.
If you want a sneak peek at the detailed technical documentation, please DM.
Contribute to protocol research - stay tuned for announcements related to this.
Contests: hop over to our Discord to stay updated with these upcoming contests: Billboards: submit clever and simple billboard ideas like you see in this post - we will be selecting multiple winners frequently and giving away real-world prizes. Name the L1 blockchain: We would be opening up the naming of the core L1 blockchain to a public opinion poll.Build stuff using our SDK Alpha Release: Coming up soon enough we will be demo-ing a game built using our SDK and making it available for tinkering around.
In the days ahead - you will hear from the Gutenberg team a lot: we have a lot of demos, talks, fun guides, essays and our detailed paper coming up for release presenting a mathematical, engineering and conceptual basis for our system and cryptoeconomics design. We have an open canvas, a lot to do still, and would love to hear from you..
No activity yet