Dialogue with Farcaster Co-founder: How can decentralized social media grow from 100,000 users to 1 billion users?

Author: Sage D. Young, Unchained (formerly at CoinDesk)

Compiled by: Yangz, Techub News

On Wednesday, Dan Romero, co-founder of the decentralized social network Farcaster, announced that Farcaster has completed a $150 million Series A funding round at a valuation of $1 billion, led by Paradigm, with participation from a16z Crypto, Haun Ventures, Union Square Ventures, Variant Fund, Standard Crypto, and others. Since becoming a permissionless social network in October 2023, Farcaster has reached 350,000 paid registered users, with network activity increasing by 50 times. This year, Farcaster will focus on growing daily active users and adding developer primitives like channels and direct messaging to the protocol. Dan Romero stated, "In the coming years, we will double down on the vision for Farcaster and truly develop it into an internet-scale protocol."

According to Dune Analytics data created by Pixelhack, on May 20, Farcaster's daily active users reached nearly 45,000, a 30% increase compared to the spike in activity on February 11 following the launch of Frames (a feature that turns posts into interactive applications). However, this figure is still far behind mainstream social media giants like Facebook, TikTok, and X (formerly Twitter).

As a leader in decentralized social media protocols, how does Farcaster plan to achieve this goal? To find out, Unchained interviewed Farcaster's two co-founders, Dan Romero and Varun Srinivasan. The two founders shared their views on a range of topics, including their upcoming decentralized plans, their current perspectives on the cryptocurrency social network landscape, and their experiences while working at Coinbase. Here are the key takeaways from the interview.

Q: Farcaster's decentralized channels are about to launch; can you tell us more about them?

Dan Romero: We believe it's necessary to allow channel creators to choose whether to set a certain economic model for the community, whether through subscriptions or requiring users to purchase specific assets for access. This flexibility can lead to higher quality output. Being a community mod is no longer a thankless job; producing high-quality content can be rewarded accordingly.

That's the advantage of channels. They are similar to subreddits, but the creator owns the channel and can manage it flexibly, increasing economic benefits in any suitable way.

Q: What is the difference between the current channels and the decentralized channels being developed?

Dan Romero: The current channels are centrally managed and stored in the Warpcast database. However, the content of the channels and casts (similar to tweets or Reddit posts) are on the protocol, so they are permissionless, while the metadata and curation capabilities of the channels are not part of the protocol. They only exist in Warpcast.

Varun Srinivasan: If you create and set up a (decentralized) channel, then all data, everything, including the channel icon and information sources, can run on multiple different clients. All data is decentralized, and any application developed can showcase the same version of the channel. Moreover, channel creators can actually control their channels, meaning if a creator wants to transfer ownership to a friend, they can. Or if a creator wants to sell it to someone else after building a following, that's also okay. Channel creators have the freedom to do so.

Q: How much effort have you put into Farcaster's decentralized governance?

Varun Srinivasan: Our decentralized governance model is very effective, known as rough consensus and running code. The IETF uses it to create all network standards like TCP-IP. It's powerful because there is no rigid structure. The logic is very simple: if you can create something useful and convince most people that it is useful, then you can release it.

There are those responsible for managing the Farcaster Hub (the nodes that store data in the network). There are those developing applications that interact with the Hub to extract and generate data. And there are users who use these applications. These three groups balance each other. … (Thus) you must launch something that most people in these three groups find acceptable. Otherwise, they will express their preferences through action. Hub operators can run older versions. Applications can choose different Hubs. Users can also choose different applications. All of this ultimately serves as a check to prevent anyone from forcing through something detrimental to the broader interests of the network.

Dan Romero: Right now, we are focused on keeping governance simple, increasing the total number of daily active users, and then launching truly useful developments. We could spend a lot of time building complex and convoluted governance structures, but that wouldn't help us achieve our goals.

Q: What has been the most important insight from your transition from the centralized exchange Coinbase to your commitment to building a decentralized social network?

Dan Romero: While people are concerned about Twitter's development after Musk took over, the network effects of Twitter are stickier than I initially thought. We entered the market in 2021, and we found that "people were talking about decentralization in the cryptocurrency space and the demand for other applications in the ecosystem." Then we naively thought, "Great, people will really want to play with new things." However, it turned out that users preferred to continue using existing products.

We really want to provide people with genuinely useful products. This has profoundly changed our thinking over the past few years… Ultimately, the only way we will win is not because of architecture, but largely because of building products that people love.

Varun Srinivasan: In the early stages, we realized that we would spend just as much time attracting users as we would attracting developers. We quickly recognized that developing applications and social products is really, really hard. Just developing Warpcast (the Farcaster client) took about 20 people a year. It’s unrealistic to expect other teams to show up on day one and decide to spend 100% of their time developing Farcaster without users.

Only by building clients, telling stories, and attracting users can we make Farcaster increasingly valuable to developers. That’s what’s most useful. Of course, this doesn’t mean building a specific SDK, toolchain, or adding a certain feature. Rather, can you multiply the number of people using this product by ten? This has been the biggest shift in our mindset since the early days.

Q: What have you learned from your experiences at Coinbase that can guide the development of Farcaster?

Dan Romero: Based on my experience at Coinbase, I firmly believe that ordinary people don't care about everything you're building. They just want a good application, and it all starts with user login. In fact, it even starts before that: how do you position the product and explain it to them?

We can do better in this regard. The current issues with Farcaster and Warpcast can confuse people. We are not even competing with fintech applications; we have to compete with the best social media applications in the world.

So once users enter the application, is it interesting enough to make them want to come back? Many cryptocurrency users like to compare cryptocurrency applications, and the reality is that if your application isn't that interesting, they will return to TikTok or YouTube.

Varun Srinivasan: Another lesson I learned from Brian Armstrong is to always focus on building what people want, regardless of changes in the primitives. He said things will never be as good as they seem, nor as bad as they seem. One reason Coinbase has consistently succeeded is that, whether the market is in a downturn or an upturn, Coinbase has always focused on output and building. We have experienced multiple cycles in the cryptocurrency space and have witnessed the implementation of this strategy; we know that long-term focus is the real winning strategy.

Q: With participants like Farcaster, Lens, Nostr, and Friend.Tech in the cryptocurrency social media ecosystem, how do you view the current landscape?

Dan Romero: Compared to centralized social media like Facebook, which has billions of users, the total number of cryptocurrency users is not significant. We are talking about tens of thousands of users across all decentralized social media (in total), who may be the same group as the 100,000 users who use cryptocurrency daily.

What we are interested in is how to gradually grow the user base of cryptocurrency applications from 100,000 to 1 billion. I don't think focusing on competitors is useful. The key question is: what do people want? What do people find interesting? If you can make users find the content interesting and fun, so they come back every day, then developers will follow. Everything else is unimportant.

Varun Srinivasan: These competitors are taking slightly different paths, while our goal is very clear: to operate in the cryptocurrency space, allowing users to build their own communities and enabling developers to create applications for these users.