Read Part 1 of this two part series to learn about the early days of decentralized social media.
Big thanks to Sismo for sponsoring this series!
Welcome back!!!! to 'The Rise of Web3 Social' series. In today's article, I'll share my findings on the Lens and Farcaster ecosystems.
I’ll analyze the two ecosystems according to a methodology that I introduce in the next section and I’ll share my opinion through the whole piece.
Three components for a successful -web3 social- protocol
I have structured this analysis following the three big components of web3 protocols:
Firstly, the protocol itself.
-
This encompasses the protocol’s capabilities and its architecture.
-
Success can be measured by the number of builders and creators using the protocol to create things.
-
It depends on protocol qualities like: relevance of features, usability, scalability and interoperability provided by the protocol (see issue #2 for the full spectrum). Builders tend to really appreciate for the core team to be responsive to their needs.
Secondly, there is the ecosystem of projects developed using this protocol.
-
This is the applications, clients and middleware tools built on top of the protocol.
-
Success can be measured by the number of products, communities and creators actively operating on top of the protocol, as well as their quality.
-
It depends developer experience, credible neutrality of the protocol, and the availability of resources in the form of capital and talent. Things that really helps builders here is an already existing social graph and a critical mass of content, good developer tools and other builders to collaborate with.
Lastly, the community that fuels this ecosystem.
-
This is a network of creators, builders, investors and consumers. Initially, the focus might be on bringing in a developer community but I don’t think there is a right sequence to bootstrap this multi sided ‘market’.
-
Success here can be measured in the number and aptitudes of individuals as well as the strength of their relationships, how often they interact, meet IRL, support each other and whether they integrate each other apps where synergies are apparent.
-
It depends on the ability of the core team to elaborate and execute a cohesive strategy.
The Lens protocol, apps and community
Lens emerged from the Aave ecosystem and core team, embodying a web3-native culture which takes roots in DeFi.
The protocol - onchain and ‘asset-first’
On the one hand, the Lens protocol was initiatlly designed with all follows and publicatons being recorded onchain in the form of NFTs, forming an onchain social graph.
By having content and profiles tokenized, one could say that Lens is an ‘asset-first’ social media (as coined by the team at Variant Fund). This makes lens social media assets like profiles and content collectable, tradable and composable with the rest of the Ethereum ecosystem. This is playing on properties that are unique to blockchains. For example one could use some valuable piece of content as a collateral for a loan to fund new creations.
Another illustration of the power of being onchain is the MadFinance app that lets brands and influencers do business around sponsored posts without intermediaries and directly onchain. Brands create a bounty for a post they’d like to see, putting the money in an escrow contract. Creators then create a post, sign it and submit it for review by the brand. If the brand is satisfied it can validate the post which will automatically be posted, at the same time the creator is paid.
Given that most activities on Lens are recorded on-chain, there is potential for a clearer and more systematic distribution of value via smart contracts. For example multiple stakeholders in the social media value chain, such as client builders, referrers, promoters, and creators, might receive a share of the fees from an NFT purchase by an end user. In the web2 paradigm value distribution was often opaque, with many stakeholder groups not even compensated for their work.
On top of that the Lens team will soon release LensV2 with concepts like Open Actions and ERC-6551 ProfileNFTs that double down on Lens advantages for composability. Open Actions let builders and creators embed other web3 applications right into the feed of Lens clients like hey.xyz. For instance I could let my followers collect a Lens post in order to gain access to joining a party.app onchain collective all in one social media post. An ERC-6551 ProfileNFT is the direct owners of all data posted through it which could give rise to interesting use cases when coupled with avatars developed by Sonar, a company acquired by the Lens team with the intention to enhance profiles. It can also be moved to another wallet.
However, the advantages mentioned earlier are partly speculations and the prospects of the protocol beyond the release of LensV2 is still TBD. In addition there may be some significant drawbacks to being ‘asset-first’ and very onchain.
It is hard to predict the behaviours that’ll emerge with financialization of all social media assets like our own profiles! Imagine an open market for twitter/x accounts. Imagine having all TikTok posts being collectable and tradable. This may lead to healthy specullation but also to poor behaviors like ephemeral cash grabs and scams. I also feel that even though monetization is a superpower for creators, financialization isn’t the fun most people are looking for on social media.
Moreover, Lens has had some scalability issues in the past with the cost per user being too high. This is the price of being heavily and tokenizing all publications on Polygon. The team has been looking to mitigate this by moving social actions offchain ,starting with publications, to a new Optimistic Layer 3 called Momoka, with the option of tokenizing the publication if necessary. It is an open question for a non-technical person like me whether this design could scale to millions of daily active users, or not.
Last but not least, as explained here by security giga brain @Oops, the current design of ProfileNFTs lets people permissionlessly squat and impersonate handles without any mechanisms in place to administer handles and protect other users from bad actors. Stani actually suggested that Oops should submit a proposal to change this at the protocol level.
A quick word on the current state of Lens protocol governance and revenue model.
Lens Governance: Currently, Lens is testing in a closed setting, and the core team controls everything. They're slowly opening up, letting the community suggest improvements through the Lens Improvement Proposals, guided by an EIP-like governance model.
Funding and business model: Lens has over $17M in funding from firms like IDEO and General Catalyst. How they'll sustain in the future is unclear, but they might adopt a fee model, like Zora and Mirror, taking a cut on value flows somewhere on their platform.
Lens Ecosystem of projects - experimenting with onchain media
Below is a non exhaustive overview of the current ecosystem of applications and tools built on top of Lens.
And usage for different apps and clients according to Zurf.social over the past few months.
The three largest applications in the Lens ecosystem are clients with a social media feed. According to this chart, Hey.xyz (previously Lenster), which is a twitter like client created during a Lens hackathon in March 2022, now has a clear market share dominance. Phaver which raised a $7m funding round recently comes second, closely followed by Orb which recently raised $2.3m.
The experience on Hey and Orb highly resembles Twitter/x besides the ability make posts gated or collectable while Phaver lets people stake on each other’s posts in order to curate them and earn rewards if they happen to have good taste.
Those observations previous to the launch of open actions which will let more sophisticated apps to be embedded in feeds. The deployment of LensV2 could trigger a wave of innovative features and enthusiasm in the Lens ecosystem.
Airstack is positonning as a leading data provider and API for web3 social, Farcaster and Lens included.
The Lens Community - still looking for healthy sources of growth
Today Lens is still in a gated phase and little more than 100k people have been able to grab a Lens handle.
Lens has had some issues with bots that it is actively working on solving through the use of AI, closed beta and potentially other means I’m not aware of. The decrease in the presence of bots on the platform might partly expain the drop in activity over the past months, not 100% sure about that.
From what I’ve seen the Lens team has been focusing its growth efforts on both developers and creators at the same time. This is in order to bootstrap the supply of applications and services as well as the precious content which is the lifeblood of all social platforms.
The content on Lens tends to be mostly created by regular users or web3 creators, we’ll see that is a pretty stark contrast from Farcasters developer heavy community. Some examples of creators are Jessyfries.lens, ChaoticMonk.lens, Grams.lens, thefaketomato.lens.
500 people are part of the Orb App “developer community” tab/channel (a group feature similar to Twitter community). It is only moderately active but probably is the best place to get updates on Lens and ask questions to other developers.
Overall the Lens community is still early and I haven’t observed a solid community flywheel in place yet.
Farcaster: decentralized social made in the Valley
Farcaster was founded in Silicon Valley by Dan Romero and Varun Srinivasan who both previously held senior positions at Coinbase.
Protocol Level - pragmatic decentralization
The cultural roots of the founders is reflected in the protocol’s elegant architectural choices and a pragmatic philosophy with regards to decentralization that their CTO calls sufficient decentralization. It states that two users in the farcaster network should always be able to communicate with each other, which implies that
-
they own their identities
-
they can post and read messages from the network in a censorship free way
Let’s see how this is done.
At a high level, user identites are recorded and owned onchain while messages and posts are stored in an offchain network of nodes called Hubs, which allows for faster processing of events and cheaper storage of user data in comparison with a fully onchain solution. In many ways we could say that Lens is course correcting towards a more offchain approach with their new L3, Momoka.
However, Farcaster doesn’t offer new sexy onchain capabilities at the protocol level today (think Lens Open Actions, collects…). Instead the FC team has focused on creating a robust set of more traditional social media features for posting, resharing commenting and so on. Therefore, an open question for me is whether this partly off-chain and conservative design of the protocol will limit FC’s ability to be a good environment for developers to unleash their creativity.
One thing that might mitigate this ‘disadvantage’ is the third party services that are already emerging in the Farcaster ecosystem:
-
indexers and data providers like Airstack and Neynar bypassing the need for an onchain social graph
-
tools like Mint could compensate for the lack of systematic tokenization of posts
-
SDKs like Discove could be the substitute to Open Actions, allowing developers to create mini apps right into Warpcast
One thing is for sure, there are no ways to collect posts, trade profiles and other social assets on Farcaster at this point.
When it comes to developer experience the only thing I can say is I’ve heard good things from builders in the FC community. The simplicity of Farcaster is an advantage in an already complex environment.
FC’s Business model: Users will pay a yearly fee to hubs in order to store their data. This is handled by the storage contract and price depends on supply and demand.
FC’s Governance: “Farcaster embraces rough consensus and running code as its governance model. Changes happen when someone makes a proposal, gets buy-in and ships running code. “
Clients and apps - developers unleashing their creativity
Below is a non exhaustive list of projects built by the Farcaster community.
Warpcast is the main Farcaster client today with 90%+ market share. It is built by the Farcaster team itself and is the most well rounded product in the whole FC ecosystem. The FC team is building the Warpcast client at the same time as the Farcaster protocol, and that they intend leaving the maintenance of the protocol up to the community on the longer term.
One advantage here is that there is one high quality client that everyone can use as the port of entry into the FC community. Another one is that the team can also experiment with new features as they develop the protocol.
On the other hand one major drawback is this might compromise incentive alignment between the core team, also developing the protocol, and other pojects competing for users, especially given that Warpcast has both large market share and early mover advantage. For this reason Warpcast might remain the main character in the Farcaster ecosystem and become a super app, a little bit like back when Twitter had third party apps revolving around it, but stronger.
This thesis can be supported through the categorization of projects in the FC ecosystem in two buckets;
-
Projects that enhance the Warpcast app: A whole range of projects focus on enhancing Warpcast features: Eventscaster for creating events, Searchaster and Findcaster searching posts or people, Launcaster for discovering projects, Discove for embedding min apps in the Warpcast feed, Weponder for creating surveys on Warpcast and so on. Making Warpcast some kind of superapp.
-
Projects that loosely integrate with Farcaster, pulling in a specific type of data: For example Fabric.xyz surfaces Farcaster profiles with a checkmark on its platform. Unlonely posts livestream transcripts as threads on Warpcast. Paragraph has integrated the FC social graph and comments directly on its web3 publishing platform. In this category the founders identify as FC community members and start integrating with Farcaster before they start adding other protocols.
It will be interesting to see which of these categories flourishes in the coming year.
Community Level - slow and cohesive growth before opening up
Note that today is Oct 10th, 2023 and Farcaster is opening to the public. This might bring significant changes to community dynamics.
The Farcaster community is very tight knit. The early community was born during the last crypto crash and was described to me as a group of “smart optimistic builders”. It is dominated by developers and founders, some create content and livestream on unlonely but I haven’t seen any full on creators.
Developers hang out a lot and love to share new experiments with each other. The current ecosystem of apps reflects this, with many one-person passion projects. I’ve been told by developers that the community felt welcoming and that the social incentives truly rewards builders that go ahead and create things.
Until today, Dan Romero has been the primary source of onboarding to Farcaster, user invites have been second. This gradual and manual onboarding has had two upsides which is slowly consolidating community culture and preventing bots from entering the platform, preserving cohesion and quality in terms of conversation and content.
Some members are really engaged and it shows in their follower counts. Most of them have much fewer followers on Twitter. Some of them raised their hands and stepped up as benevolent channel (opt in feeds with various topics) leaders on Farcaster. A few really active users account for most conversation liquidity on the platform.
Now what will happen after Farcaster opens up is an open question. It should face the same issues that Lens has faced regarding bots and spam. And its initial community could also loose the quality of being a trusted and comfortable space.
The research continues…
Thank you for reading!
At this point is clear that Lens and Farcaster are very different by design. While I have more theoretical affinities with the composability and web3ness offered by the Lens protocol, I admire how much the Warpcast team has achieved building a protocol, a client and a community all at the same time with high levels of success so far.
Will LensV2 bring a new wave of innovation to the Lens ecosystem? Will Warpcast keep rising to dominate the Farcaster ecosystem? Let’s check in on these questions in a few months.
I’ll add both FC and Lens projects to the mapping I have first shared in issue #1, and we’ll keep an eye on these two protocols as we explore the identity layer of web3 in issue #5.
Resources
-
Article explaining Farcaster’s approach to identity and data
-
Detailed article on Farcaster By LJW and Cameron Armstrong
-
Sufficient Decentralization by Varun