The internet is at a crossroads.

Web3 is on the verge of mainstream adoption, but it still lacks a use case compelling enough for the general public to drive a switch over to it from the current system. At the same time, public distrust in private control of the internet has been growing. Major tech companies like Facebook, Google, and Twitter collect and monetize user data, leveraging their control over algorithms to shape public perception and the flow of information. People are increasingly aware that these centralized entities are monopolizing access to the core value the internet provides society.

These two societal currents are converging towards a natural solution: a digital commons owned by all participants - our Third Space. This term extends the concept of third places, as a social environment separate from our home or workplace, to the digital realm. 

Beyond the financial applications, blockchain technology can also be a powerful tool for tracking participation, contribution, and power in a community. It is here that web3 will break out its holding pattern of bubble and burst, to provide a more consistent and essential value to the world. We can leverage these tools to make social media that is for the people, by the people, and that distributes the value it generates across the people who created it. This ethos is in stark contrast to the traditional model of extractive social media. It is with this new ethos that we will build a movement.

However, there is a lot of distrust to overcome. Between environmental concerns (which proof of stake networks have largely solved), and pump and dump schemes (which the community is getting better at catching), blockchain technology has garnered infamy among the general public. It is a hard reputation to fight against when considering onboarding lay people into this world.

This leads us to the most important question:

How will this time be different?

The beginnings of what a Third Space could look like have been started with the Farcaster network. Farcaster provides an amazing backbone to power social media, offering a Twitter-like posting and following service that can reach large audiences with a robust system architecture. At its core, Farcaster is a distributed data source. It acts as a layer of data for the internet that any social media client can access and host, replacing the centralized server that Twitter or Facebook keeps.

Farcaster clients (like Warpcast) act as the user interface for this social media system, providing a window that can be focused onto different sections of the Farcaster data layer (with all clients accessing the same fundamental pool of data and social graph). A similar system is being built by the AT Protocol and Bluesky. But clients for both of these decentralized data systems have set their ambitions at rebuilding Twitter, instead of innovating on the experience. 

We can take one small step (or perhaps a giant leap) further and build a client that empowers users to craft their own way of experiencing social media, a personalized vessel for navigating this shared layer of online content and apps. Rather than a one-size-fits-all approach, users and organizations can create and share social media spaces with unique themes, layouts, and data sources. Communities can form around specific topics and curate a shared social experience with tailored feeds and discussions centered around their interests. Imagine that this platform can then also seamlessly integrate different Web3 experiences and information into a single ecosystem. Let’s call the technical system that enables this the Space Protocol and the client built on top of the system Nounspace. 

Let’s paint a picture of what using this platform might look like. After a stressful day, you open Nounspace, seeking a break from the crypto noise and wanting to unwind with some light-hearted content. Within the app, you have personalized feeds tailored to your interests, called spaces. You navigate to the cat space, a curated stream of adorable feline photos and videos, providing the perfect escape. Once you've calmed down, you feel like engaging in something more interactive. 

Without leaving the platform, you switch to a gaming space, where you can access a popular free-to-play fighting game. The game's cosmetic items are stored onchain as NFTs, allowing you to truly own your in-game assets. You dive into a few matches, setting a new win streak, and the app prompts you to share your achievement, automatically generating a cast for you to post to the built-in Farcaster feed.

Feeling reinvigorated, you move over to your personalized news space to catch up on the day's events. Throughout this entire experience, you never left Nounspace, seamlessly transitioning between different Web3-powered sections and activities – from consuming lighthearted content to gaming with owned assets and sharing achievements, all within a single, unified platform.

Nounspace leverages Web3 to create a cohesive, user-centric social media experience tailored to your needs and interests. It empowers you to own and control your digital assets and identity, while providing a sanctuary from the noise and manipulative feeds of traditional social media; offering a range of integrated and personalized experiences under one roof. But most importantly, all of these experiences are part of an open ecosystem, not a walled garden. Anyone can contribute and make building blocks for these spaces. We call these blocks Fidgets.

This section will get a bit more technical as we dive into the long term vision for Fidgets and the Space Protocol they are built upon. This is not intended to act as a full technical specification - we intend to release a white paper when we get closer to public release. Dynamic feed choices is just one example of the creative power that the Fidget system unlocks for users - the architecture defined here will enable a wide range of applications and experiences. 

Fidgets are the essential building blocks of the Space Protocol. A Fidget can be thought of as a self contained app. Fidgets can be installed and arranged onto grids known as Spaces, which serve as a dashboard for social media experiences.

But Fidgets are much more dynamic than an app. Fidgets are fully customizable and composable web experiences; anything from a game to a meme to a part of another Fidget. The entire architecture of Nounspace is Fidgets, all the way down.

You can nest any Fidget inside of another Fidget. Someone could build a music player Fidget, and someone else could install that into a Playlist Ranking tool. If you feel like you know how to make a better UI element, you can fork the existing UI, and replace that button you hate with one that you love. If you know how to code, everything about a Fidget is customizable. If not, those that do know can include customizable settings, like colors, fonts, or the contract address you want the Fidget to query.

Initially, core development will target small, simple Fidgets. These will act as building blocks for more advanced and interesting Fidgets.

A complete Fidget can be installed and run by a user inside of a Fidget Virtual Machine (FVM). It has everything that is needed to render the Fidget in the user's browser or native app, and run it. In contrast, a partial Fidget requires other components to be completed in order for the user to interact with it. Examples of some partial Fidgets include a data API to access FC data or a design style for a button.

Complete Fidgets must perform tasks across the 3 layers of the Nounspace system. Each layer handles a different essential aspect of the social experience: the data layer, the algorithm layer, and the display layer. Each layer can be customized by the user.

How data is stored and accessed. This is both personal user data for the Fidget, but also general information from public sources. A complete Fidget needs a way to write to data. It can store this data on IPFS, Farcaster, AT Protocol, or any other service. The Fidget that manages access to the API of one of these services covers the Data Layer

The way data is selected, and ordered for the user. This is the source of actual decision making power for the Complete Fidgets. These will access data from the Data Layer and tell the UI what to display. They also handle user input and pass relevant data that needs to be stored back to the Data Layer. 

How the data is displayed. Everything that falls under the category of UI and graphics. In the case of a social feed, this layer could determine if the posts were Twitter or Reddit styled. It also acts as the input layer for the user to manage the Fidget settings.

The way the user engages with the system, and manages what experiences they are using. Inside a client will be an FVM, running each of the Fidgets that a user installs. Nounspace will provide an open source client for public consumption. This client will come pre-installed with a basic theme for the user's initial Homebase and a Fidget discovery tool.

The most comprehensive example for a complete Fidget is a social feed. Let’s look at what a core Farcaster Feed Fidget would look like. 

The data layer will be built as an access point to a hub. Nounspace will use Neynar to provide this, but it could come from a self hosted hub. Whoever develops the Fidget can leave the configuration for how to access the hub. It would also be a valid implementation to have the data layer be built on top of a SQL replica of the Farcaster network (the type that hubble replicator creates).

The algorithm layer will be a simple “most recent” feed. It will query the data layer for casts, ordered by timestamp. A more complicated algorithm could be fed user data about likes to create a “most interesting” ordering instead.

The display layer will be formatted like a Twitter feed, with simple block dividers.

To enable customization of the UI, we will need to introduce another data layer Fidget that will talk directly with the display layer. Since Farcaster currently does not support large amounts of profile data being stored on the platform (April, 2024), an IPFS connection will be used to write unsupported settings data for the current user and this specific complete Fidget. An IPFS connection Fidget will be used to manage reading and writing this data.

Initially, Fidgets will only be developed internally and the options for where to source Fidgets from will be limited. Once we have reached the point where we can support allowing anyone to publish their own Fidgets, we will launch a Fidget Registry to track all Fidgets and their versions and forks. This registry will be funded as a public good. The Fidget Registry needs to exist as a distributed source of information about which Fidgets have passed the reviews to make sure that they are bound to the right underlying smart contracts, and not an attempt to scam developers or users who install them. (This also helps prevent a situation where an open source Fidget is used to power a for profit Fidget, but the for profit Fidget does not properly set up the contracts to pay out the open source Fidget for its use)

One of the biggest issues on the internet is how we pay people for the wonderful work that they do. Web3 has been largely working to deal with these problems and offers some of the most transparent and community driven solutions. We plan to follow the principles outlined in Morpheus' MRC21. If we treat each Fidget, complete or partial, as an NFA (Non-Fungible Application), we can track who owns each component part of the complete Fidget. Any revenue generated by an NFA will be divided among the subcomponents, thus sharing value with anyone who works on the development of a Fidget, including open source developers who write code for the good of all.

A small portion of the revenue generated by any Fidget will be streamed back into Nounspace. Nounspace will share these earnings through tokens which will be distributed to the following groups:

• Community, for the value that they bring to the network & the connections they create.

• Developers, for the Fidgets they develop and offer for free.

• Protocol, to fund ongoing development for the open source Space Protocol code that powers all of the Fidget ecosystem.

• Team, to maintain the Fidget Registry and the publicly available free Nounspace client

• Capital Providers, for their support in making this project happen.

All Fidgets in the Nounspace universe will be free to install and configure, they just might not function properly until a payment is made and recorded on a blockchain. It is easier to gate at the app level than it is to gate at the installation page. This gives every developer much more control over the costs of their app.

This means that installing the Fidget is free, but it can check the chain for a token saying that you own access to the Fidget and gate full functionality if you don’t.

The beauty of this system is that it allows all of the ownership of apps to be recorded in such a way that you will always be able to install anything you own again, onto any new device you get.

Lay folk who are not well versed in web3 will largely be turned off by a platform that seems to be a hot bed for everything that they have been told is bad about crypto. It is going to be hard to break out of that world. 

To do this, we have to fix three problems for these folks:

• Reduce the technical overhead and make it easy to understand
  This means things like: make the sign up flow feel like any other web2 website, no secret recovery phrases. Onboarding needs to start in a web2 like manner and slowly introduce web3, not slam them over the head.

• Make the platform free to use
  Requiring a user to invest any money, whether that is for a membership, or to buy tokens, will be a dealbreaker. New users must be able to sign up and be off to the races without spending a dime. Money can be made from users down the line, but the core features of the social platform must be free.

• Make the content broader than just degen activity 🎩
  Excessive degen activity will turn off ‘normies’ that are new to web3. We can only push their comfort zone so far. We need content that is relatable beyond the web3 bubble. 

To make web3 accessible, it has to masquerade as web2. Always meet your users where they are at. No one wins if you make the end user feel like an idiot. There are a couple ways that we can achieve this, but one is going to be a real deal breaker.

We have to abstract away the wallet by embedding it. This is already catching on in other parts of the web3 world. Requiring users to set up and manage a wallet is not going to work. So how do we address this? We make the client and the wallet into the same tool. If the client has a basic sign up and login flow using a username and password, the web2 users will feel right at home. Under the hood, the client will create a key pair and encrypt the private key online, using the user's chosen password. Very similar to how a password manager stores data. This will allow users to login to the client from any device and access their same account. 

Since under the hood it is all powered by wallets, there will still be ways to create new accounts and login using other wallets, but these are for trained wizards only, not muggles.

Every social media platform carries a large number of people who exist on the platform, but who only participate through likes, if at all. Reddit has named this group of users Lurkers. Lurkers are not a bad thing, as they make up the 80% of the population that consumes content, compared to the 20% that create content (these are not study backed numbers, 80/20 split is just a famous one that is probably not far off). 

Lurkers want to consume content, and generally want to do so without spending any money. Can't blame them! But how can a world powered by paying content creators for their content appeal to a general public that wants things for free?

Traditional social media has solved this problem with ads. Charging their users through visual clutter. I am not opposed to ads, but I am opposed to having them shoved down my throat without choice (I use an ad blocker for a reason). 

What if, instead of ads, people who wanted eyes on something could offer to pay users for their time?

We make lurkers into contributors instead of detractors by allowing them to be paid for their attention. An advertiser could pay the user who chooses to watch their ad, a researcher could build a poll and pay users to engage with it. They in turn can take the money made and give it to the content creators that they want to support. We create an in app ecosystem that allows even the cheapest schmuck who does not want to spend a dime to still earn rewards. Then they can take these funds and go tip, purchase fidgets, mint NFTs, and participate in the web3 economy that powers Nounspace.

Finally, any social media platform has to be just that, social. If we only build an insular community of web3 nerds, web2 normies will not want to get on board. This is where the dynamic data layer comes into play. We can build Fidgets that pull from any data source. Most importantly, we can merge into other decentralized social networks, including more mainstream ones like AT Protocol. This will allow Nounspace users to interact with folks outside of the Web3 ecosystem, like the users over on Bluesky. And it will allow existing Bluesky users to easily switch over to the Nounspace experience.

The larger the network graph that Nounspace is connected to under the hood, the more appealing the project will be to a broader audience.

Any good product needs to start with focus and direction before trying to broaden its reach. Our first step for Nounspace will be a Nouns targeted space. The Fidgets in this space will serve as an example of what can be built and to act as a proving grounds for the Space Protocol. This means that the initial tooling will be built to target the following data layers (this is not a comprehensive or complete list for the long term):

• Farcaster

• Nouns 

• Etherscan

• Coingecko

For items in the other layers, Nounspace’s core website will act as the platform layer. We will build initial versions of algorithms and feeds for Farcaster data. More advanced implementations will roll together different data sources into a single feed. 

As we develop, we want to gradually extend and develop tooling for communities beyond Nouns. These will start as recycled versions of previous tools, adopted to fit a more generic profile (i.e. adapting the Nounish Governance Fidget to support SnapShot spaces).

At the same time, we will also begin to tinker with a broader range of customization options and theming. Once we feel comfortable with the API for Fidgets, we will open up the floor for community submitted Fidgets. Initially, we will perform a manual review process for Fidgets to make sure no malicious actors are taking advantage of the system and ensure that we have ways to catch them before we open up Fidget development to be totally public.

While Farcaster is the primary focus, in the future the core team will focus on developing open source data layer Fidgets that will connect to other networks beyond Farcaster (AT protocol being the obvious choice, though Mastodon’s ActivityPub would also be a ripe target). There will need to be some adaptations made to the signup flow to support managing multiple different accounts. A well made data layer Fidget will allow developers to onboard and login users to multiple different platforms. By offering connections to other social media sources, the Space Protocol will be able to expand and become a universal protocol to unite disparate social data protocols into one experience.

We hope you will join us on this journey towards making the internet a more collectively created and owned experience.