On September 17, FarCon Asia, Asia's first large-scale Farcaster ecosystem offline event hosted by Takocast and co-organized by BlockBeats, was successfully concluded. At the conference, Farcaster KOL Ted had an AMM with Farcaster founder Dan. During the conversation, Dan emphasized the differences between Farcaster and Web 2 social platforms, and explained why users are encouraged to speak as individuals on Farcaster instead of creating brand accounts. He also shared Farcaster's developer ecosystem and how to provide developers with rich tools and opportunities through permissionless social graphs and content distribution. BlockBeats organized and edited the text version of the content of this meeting, hoping to help social application developers and audiences. The following is the content of the conversation:
Ted: Could you give us a quick rundown of why you created Farcaster, what Farcaster is, what Warpcast looks like on top of Farcaster, how it’s developed so far, and then we can dive into what’s next.
Dan: I'm Dan Romero, one of the co-founders of Farcaster, and I've been working in the crypto space for ten years this year. I've been working on Farcaster for the past four years, and before that I was an early employee at Coinbase and was responsible for leading Coinbase's international department, so I used to travel all over the world for business, including Singapore.
Farcaster is a decentralized social network, which is an underlying technical system that allows anyone to access the social network without permission. In simple terms, this means that the social graph (who follows whom, what people say, and how people respond) is freely accessible. This is completely different from the social network model of Web 2, such as Twitter or Instagram.
We are also developing Warpcast, which is the initial application based on the Farcaster protocol. I guess a lot of you are already using Warpcast, or you started with Warpcast and then tried other similar applications. This is what we have been doing for the past four years: the protocol and this initial application.
Dan connected to the conference site
We've had significant growth this year, especially this year, which has been very impressive. We have about 700,000 registered users, all of whom are paid. Users register on the Ethereum L2 OP mainnet, and the rest of the system runs off-chain. You can understand it this way: once you register, it is equivalent to purchasing a ticket to participate in this system. You can use this system at will, and the data is stored on nodes around the world. These nodes are called hubs.
We are currently focused on two things. The first is growing the number of daily active users in the protocol, which is mainly achieved through our ongoing Warpcast application-related initiatives. The second is another set of initiatives around the protocol layer, which is to enable developers to use the system development layer to build any application they want, gain access to the social graph, and reach the growing user base of the Farcaster protocol. Overall, we are working to make it easier for others to develop applications with social features, all based on crypto APIs and infrastructure.
Ted: Regarding the developers you mentioned, obviously Tacocast is a client built on Farcaster. We just had Jesse talk about frames, can you talk a little more about the developer building blocks, what these "Lego blocks" are, and what you think about the developer ecosystem? Especially with quite a few projects and developers here today who are also using these developer blocks to build their own projects?
Dan: From a developer's perspective, Farcaster can actually be broken down into three tiers. The first tier is where you get started with the system, which is a giant permissionless dataset that you can access directly, run a hub, or use an API provider.
When a new user signs up with their crypto wallet, you can use these APIs to check the address and get a ton of data completely permissionlessly. If the user has a "professor" level profile, this can make your signup experience magical. You can pull their profile photo, bio, and even find out who they follow on the web. More importantly, not just who they follow, but who they interact with most frequently. This way you can almost instantly create an onboarding experience that shows them their closest friends.
Another way to do this is to recommend who they should invite to join the app based on network usage. If you think about Web 2 social networks, one of the important factors that helped the growth of large-scale mobile apps was the contact API. For example, when you sign up for a new app, share your contacts, because by matching phone numbers, you can quickly find people who are already on the app and recommend you to follow them.
These apps also have the ability to tell, for example, that 37 people have signed up with Dan's number, and then prompt others to invite Dan, because we already know he's likely to show up. The most likely reason to continue using an app is that your friends are also using it. So this is the first way to integrate with Farcaster, and it requires almost no special work. You just build your app and make sure that each Ethereum address registered can be checked by Farcaster.
The second level of the Farcaster developer experience is to use Farcaster for content distribution. This is very important. Most crypto applications are spread through Twitter or other similar social media. Sharing your website link through Twitter, such as minting NFTs or downloading applications, the reality is that Twitter's algorithm now severely limits the spread of tweets with links.
On Farcaster, you are not restricted in posting links. We have even designed a set of features to enhance the spread of links. You can think of it as the simplest operation, posting a URL. Now, most modern applications, if this URL is a link to a news website, will use a feature called Open Graph to automatically pull images to make the link preview richer, thereby increasing the click-through rate. Farcaster uses the same concept and continuously enhances the function of links. You post a link, add an image, and then Farcaster will add buttons, and even start transactions directly in the dynamic, providing some basic input and output operations. This is the so-called "frames", which is actually a functional enhancement of the link.
I think the third layer of development experience on Farcaster is to go beyond the permissionless use of the social graph and content distribution to further develop more Farcaster data capabilities. For example, an app like Tacocast, not only can you do everything you can do on Warpcast, but you also get a bunch of additional features. They have a great translation feature, if you are not a native English speaker or even do not understand English, you can use this app to automatically translate all the English content on Farcaster.
Project owners are not encouraged to create brand accounts, but to speak directly as individuals
Ted: For everyone here, can you give some advice on how to get started and how to build your own brand?
Dan: I think authenticity is very important on Farcaster, and people care more about users they feel are genuine than on most social networks. So some bad habits that you bring from other social media are not welcome on Farcaster. Some people may be confused, like "I use this on Twitter, why not on Farcaster?" or they don't quite understand the culture here.
But in fact, if you are building a project or company, Farcaster usually does not encourage you to create a brand account. It is more about you as an individual, as the creator of the project you build, speaking directly as an individual. I rarely use the official accounts of Farcaster or Warpcast myself. Whether it is on Farcaster or Twitter, most of the time I speak with my personal account, which makes the interaction between people more personal and resonates more with the early community.
So I think the first point is to be yourself, which is much more important than following those so-called "10 social media tips for building a brand account." Just like when you go to a party, introduce yourself first and then introduce your project, instead of just introducing the project and making yourself irrelevant.
We have a feature called Channels that allows users to show multiple sides of themselves. If you come to Farcaster and only focus on showing one thing and don't care about other content, it goes back to the issue of authenticity. People will feel like you are just here to "drop something off." So take some time and don't focus too much on being a megaphone, but focus on interacting. Even if there are only 10 people following you or 100 people, it's okay if you actively interact with them.
If you can really start building a community on Farcaster, it can be very powerful. We have a lot of examples of people who have never been successful on social media in the past, or who wouldn’t consider themselves a social media type of person, but because they were authentic on Farcaster and it resonated, they were able to build an audience.
The last point is that some people may feel that their English is not good enough or that the network is too US-centric. This is true because the core team is in the US and so are many early users. But I would say that first of all, Farcaster is a decentralized network that is now used by people all over the world, and the number of people is increasing.
Secondly, I don't think everyone needs to interact with everyone else. Some people may misunderstand that Farcaster will be like the "United Nations" where everyone talks to each other. But in fact, it is like other social networks. You will naturally see various sub-communities and different small circles, and people may be completely unaware of what is happening in other circles. It is no longer like a big group chat, but more like a diverse world.
I think that's a shift that we're still going through, especially as we grow, where it used to feel more like a group chat, but now, especially with the idea of cozy corners that we're launching soon, you don't have to interact with the people with the most followers on the network. You can find a niche or a group of people who share your interests and they can grow completely independently and not care what people in California think, and the network is still there for you.
I think that mentality is important and sometimes people focus too much on the fact that Farcaster used to be a bit biased towards the US, but that is not our goal and I am sure that in the next year or two we will make a lot of progress in that regard.
Ted: Speaking of "comfortable corners" and your next focus, perhaps slowly shifting to a decentralized model, can you give us a brief introduction to the future direction of Channel and where you will focus your investment?
Dan: For those who are not familiar, there is a concept called "Channels" on Farcaster. Today it is a bit like a hashtag, but the difference is that you can only use one at a time, unlike Twitter or Instagram where you can have many hashtags. And these channels have a certain management mechanism. Someone is responsible for managing the channel and deciding what content is allowed and what is not allowed. Of course, you don't have to post content in the channel. You can only post to your followers, or create a channel yourself.
Channels are carefully curated spaces. For example, some people may go to the topic channel they are interested in after get off work to participate in discussions, but they are not willing to spend a lot of time managing content. I think people are not so sensitive to content management, but they are more sensitive to the management of "people".
It feels more like a club or a group that you're a part of and you can create a shared identity and really build a community, which I think is one of the biggest shifts. And of course, we're also constantly improving other features, mainly giving more tools and more control to the moderators of the channel and the people who run the channel.
I think it does create a stronger sense of community naturally because you have fewer people you actually know. While channels can get larger, maybe 1,000 or 10,000 people in a group, I imagine the average channel size will probably be more tight-knit, 50, 100, 200 people who can post content, but the sense of community will be very strong.
Questions from the audience
Question 1: For the next phase of Farcaster, especially in Asia, people are looking forward to seeing growth. In the early days of Farcaster, the $5 registration fee was to prevent robots from joining, and I remember we also saw some robots. So my question is, what do you think about removing this fee to make it easier for users from other platforms to join Farcaster and participate in social activities?
Dan: We think about this a lot. First of all, if you run a centralized network with pre-set software, then theoretically anyone can register from anywhere through a contract on the chain. If this is free, it could bring the entire system down. The reason why Ethereum has gas fees and Bitcoin has transaction fees is that if there are no fees, the state of the network will increase significantly, and people may send a large number of transactions to disrupt the network, which will have a huge impact on the ability to remain decentralized in the future.
So we won't remove the fees completely, but we can consider other ways, such as we currently have about 700,000 users, and maybe 7,000 of them are very trustworthy. These users can get 10 invitations through their on-chain activities, or through some kind of "human proof" mechanism. Add new invitations every once in a while, and grow the user base more systematically in this way, while greatly reducing the risk of robots and spam.
I think what we might do is lower the price, and we've actually adjusted the price several times. At the protocol level, it's the equivalent of $2 in Ethereum right now, and Warpcast is $5 because it's an in-app purchase. That's working fine for now, but we might get that down to $4 at some point. If the fees at the protocol level go down, we can continue to lower the price, but there's always going to be a certain amount of fees in the system because you have to prevent spam.
Another measure we may consider is to more actively give more permissions to some high-quality users who have contributed to the system, published a lot of content, and had good interactions with multiple parties. This is also a decision we will make intentionally.
Question 2: The usage of frames has dropped significantly recently. How do you think frames will perform in the future? Will their adoption improve?
Dan: Yes, overall network usage has dropped recently. Initially there was a big spike in usage due to the introduction of frames, just like any social network has a spike. Most of these spikes are temporary, and then you need to think about how to build on them. I think this is one way to understand the usage of frames.
Farcaster protocol DAU/MAU ratio, which can measure user stickiness
Secondly, the current frames function is still relatively basic, based on pictures, optimizing the experience of operating in motion. For example, if you click the "Floppy Caster" button in the middle of the bottom of Warpcast now, you can play a complete game in the app. This is an experiment, and you will see more similar attempts from us. In other words, frames are no longer just simple pictures, but provide users with a richer application operation interface.
Dan: The second use case of frames, which I think is also a relatively high-frequency and well-received feature, is to initiate transactions. For example, when posting a link to an on-chain transaction on Twitter, at most a website preview can be displayed. Users need to click on the link and connect to their wallet, which often encounters problems when operating on mobile devices.
In frames, you can directly initiate transactions based on the context of social dynamics, whether using NFTs or exchanging tokens. This allows users to use interactive functions more conveniently in the current environment, whether it is a complete small application or guiding users to perform on-chain transactions in a safe way. We are optimizing the speed of transactions to make transactions faster and smoother, reducing steps. Now you only need two or three clicks instead of cumbersome operation processes.
Question 3: What can we as Asian audiences do?
Dan: First of all, thank you very much for hosting the Farcaster Asia Conference. I don’t know if you know, but the founders of Tako flew to Los Angeles two weeks ago just for one day to introduce me to the Farcaster Asia Conference. I was really shocked by this support.
I think for those who build applications outside of the United States, they will have a better understanding of the local market or regional needs. A good example is our translation function, which can serve markets such as China and Japan. So, I think the most important thing is to create some communities on the network that are not necessarily geographically centered, although there will be overlap and interaction between these communities, but ultimately this will have a huge impact on the overall development of Farcaster and service providers.
Question 4: I would like to hear your views and vision on combining social growth with financial applications. Since most products focus on building social outputs, I would like to know how you view using social data to provide users with more social financial experiences?
Dan: I think when you have a permissionless social network, you can really get cutting-edge data, which has multiple values for developers. Another feature of Farcaster is that in theory everyone has an Ethereum account, even like Nostr, which uses a slightly different public and private key system.
As far as I know, there is no such mandatory requirement for decentralized Twitter alternatives, except for projects like Lens or a few others. I think this is very underestimated, we have seen some permissionless features and tokens like Degen or Moxie, this deep crypto integration makes all this possible, you can do it directly on the social network without complicated operations.
A small example is that we started offering a $5 USDC reward to the top 100 users of the day, which is a bit like Elon's distribution model for Twitter ad revenue, but we are a small amount of money settled daily. We will release this feature in a few days, and if you think about other payment methods, you may encounter 100 different problems and only get paid at the end of the month. Or PayPal accounts are frozen, unavailable in certain countries, and there are countless other things that can go wrong. With USDC paid to an Ethereum address that you control, as long as you don't lose your keys, there will be no problem.
