This post is taking part in the Farcaster 2026 writing contest

Communities

In the web3 world, when people hear the word community they might find themselves thinking about their favorite project community, whether they call it a DAO or not. Or it might be a creator community, or an ecosystem community. Regardless, it probably exists on some platform like Discord, Telegram, or maybe something decentralized like Farcaster. The number of people that fall into this demographic, however, is very small.

The rest of the people are still on Instagram, Facebook, X, even Reddit. There is currently a migration happening towards more decentralized options, such as Bluesky or Mastadon. These communities are much more diverse, ranging from hyperlocal communities to ones that span the globe. Topics that unite these communities could be anything from the kind of vehicle they like to what kind of food they eat. Topic-based communities make up a significant amount of online engagement. Even in Farcaster now you find more and more people migrating to group chats or channels, instead of engaging in the main feed. TikTok shook up the whole industry with a new tactic: dynamic communities. They let people's own behavior form dynamic communities around algorithmically distributed content, instead of the more static foundation of self attested interests most web2 products started with.

This natural tendency for people to congregate around topics of shared interest is the very fabric of communities that has existed far before technology ever existed. It is the natural human phenomena that drives the persistence of social media in the modern age.

The Next Evolution

In order to move past the dark loop of the attention economy that currently powers web2, we have to provide something to people that they are missing. This isn't any one thing, but rather it has to be the right combination of things in order to attract them.

The two main narratives that seem to have struck a chord recently are:

1. Decentralization

   Everyone is growing increasingly disillusioned with centralized social media. Whether it's because the government was secretly in cahoots with media companies, running a censorship campaign-- or because now the social media companies are the government, there's plenty of reason for any side of the fence to be severely concerned with the risks of centralization around any media, social or otherwise, and not just in the US.

   The global battles of the future will be fought over data ownership, not land.

2. Creator Compensation

   This one might seem obvious, but it's also surprisingly complex. Obviously creators want to get compensated. The ability to monetize one's digital work is the driving force of the economy in an ever increasingly digital world. The complex part is that people are picky about how they engage with this monetization. The majority of the world is far less interested in tokens and incentive schemes rooted in cryptography and mathematics. They are comfortable getting paid to fill their content with ads simply because it only steals from them in ways they've already grown accustomed/numb to.

   Ad-driven creator economies only have to prioritize short attention spans. They don't have to go deep into a subject, worry about fact-checking or corroborating research, or even have any conviction around their content. It simply must grab people's attention for long enough to deliver them an ad.

Whether its entertainment, productivity, or personal growth, the world has turned to social media as the hub of their engagement with the world. This leaves us in a battle for Mindshare.

Capturing Mindshare

It seems clear to me that most people outside of crypto natives do not trust the crypto ecosystem, whether despite of or because of the current public news around it. Every cycle acceptance grows a little bit more, and retains adoption a little bit more, but mass adoption doesn't happen because as soon as anyone sees behind the curtain even a little bit, it's obvious the emperor has no clothes. At this point most leave, but some do stay and build value driven projects-- or they stay for the casino and never leave. Some people live for the casino.

The casino of crypto will continue to churn through the waves of adopters, and time and time again it ends the same: most users disillusioned and distrusting. Since the casino can drive enough narrative to build momentum for those at the crest of the wave to cash out before the crash, the cycles keep going. People can't help themselves but try to catch the next wave. Somebody's a winner, and next time it'll be me! But the people building the value-driven products aren't the ones winning in this scenario, ever, so the real products that actually serve people's needs don't get built. Just more games for the casino.

True adoption comes from remembering what drives communities in the first place:

Shared interests and values.

This is the secret formula to winning the mindshare in the great migration happening right now from centralized to decentralized platforms. Focus on optimizing for communities to share their interests. Drop the crypto, drop the tokens, and drop the speculative incentive schemes.

Focus on helping people who have specific interests generate high quality content about that subject to share with an audience that will value that contribution and have a productive response to it.

This is the necessary foundation before you can start monetizing anything. Farcaster proved that in seeding the initial community, and now it's time to do that for other communities.

The Needs of Existing Communities

Whether they are unhappily located on a different platform, or actively interested in finding a new one, there are very few people that would resist trying something new in the volatile moment we find ourselves in-- with the exception of influencers who have actually achieved a following on centralized platforms and already produce a legitimate income.

This is a perfect opportunity to onboard new users en masse.

But the trick isn't in getting them to try it. The trick is getting them to stay. There has to be significant value for a user to shift their attention space from one platform to another, even if they're already interested in doing so. The web2 model is to maximize for convenience and dopamine. This strategy has effectively killed the entirety of the traditional forums prior to the social media age. Almost every community that existed on a forum has migrated to some kind of social media app that uses algorithmically delivered content that maximizes for short term attention, so the individual has to cycle to a new piece of content quickly so there can be another opportunity to inject ad space. By maximizing for instant dopamine release, this Attention Economy is able to gain surprising user retention. It's the perfect combination of factors to keep people engaged: rage bait, personal validation, and the promise of rewards if you win in pvp against your fellow community members. The easy escape to games, short-form videos, and memes of cute animals keeps people desensitized to the loss of rich, deeply engaging content.

Dan Romero on Warpcast

Have you tried the onboarding flow recently? What % of casts are crypto posts if you don't select Ethereum as an onboarding interest. The issue is overall supply of interesting and entertaining content. If we can 10x that, may be enough to keep people around. Revealed preference is people want infinite scroll (why Twitter For You, TikTok, IG Reels do so well metrics wise).

Farcaster

yesyes on Warpcast

Pretty much this. I think the "revealed preferences" stuff applies in some cases but i believe even @dwr.eth uses them too much. Example - assumption - "Unhealthy person states that they want to eat healthy but continue to eat unhealthy food so their revealed preference is that they want to eat unhealthy" possible reality - Unhealthy food is more readily and cheaply available than healthy food. If healthy food was present near to the person then maybe the person would at least start eating somewhat healthy. https://x.com/nearcyan/status/1812028627690917966

Farcaster

By over-focusing on the 'revealed preferences' of users, we've lost the focus on what creators need to create high quality content, and what communities need to showcase that content. All users are revealing w/ their preferences in apps designed around maximizing short-term attention is how best to play to the metrics and maximize ad revenue/attention capture.

This doesn't help communities grow.

So what are the actual needs of communities?

Somewhere in the transition to modern social platforms, we lost the focus on quality of content and depth of discovery. The flash and payout of the short term attention economy buried it. While platforms serving ads might be more successful with these tactics, communities suffer. This results in quality contributors burning out and the churn of low quality contributors continues on while the influencers that stick around profit by continuing to pump out content that fits this short attention span model, with no increased depth or exploration of the subject matter.

This is resulting in the erosion and disillusionment of many previously thriving communities. These are the characteristics that I see missing in the web2 social media world that was either lost from the previous era of forum based platforms, or was not improved upon and remains an opportunity space:

1. Onboarding

   New users are just thrown to the wolves. Having onboarding options for communities that can be custom tailored to the needs of the community is crucial. This is a space that AI can serve, but only if it's seamlessly integrated in way that provides a positive experience, not a frustrating one. How the AI is trained and implemented becomes critical-- a design space that needs experimentation.

   Data ownership is a critical factor here, and one that blockchain provides ample opportunity to tap into, but we haven't seen the practical application of data ownership in a meaningful enough way to create value in this opportunity space. Community owned AI tools is low hanging fruit here.

2. Curation/Moderation

   This was the secret weapon of Reddit that made it dominate early on in the development of web2 apps. The up/down vote was such an effective curation tool that it drew a massive number of people into it. The exploitable nature of it, however, led to the only retention being a very specific brand of person that needs no further description. This is an opportunity space that has not been improved upon yet. To this day, many people still rely on Reddit to find social proof of questions they have, and many LLM's in the limelight are trained off of this data due to its unparalleled value.

   This is somewhere that platforms need to focus on flexibility, not rigidity. Create an environment for communities to experiment with different curation and moderation techniques, tools, and strategies. Let distributed experimentation find the solutions faster than anyone could arbitrarily design them.

3. Knowledge Management

   While connected to onboarding, this is one thing that was actively lost when migrating from forums to web2 based social media. There is absolutely no consideration for knowledge management for communities in the algorithm. There's little room for injecting ad space if people are able to quickly, easily, and conveniently get the information they need and get back off the app. This has resulted in the design space catering towards maximizing people's attention on the lowest value content: repeated basic information.

   Instead of encouraging the gradual deepening of collective knowledge around this shared topic, the community devolves into basic arguments over simple facts that are often misunderstood because of lack of context. The context is never developed because the app is designed to push people to new content to engage with so they can see more ads. Staying in a discussion and going deeper could be used to update a knowledge repository that is used to enrich the onboarding experience for new members, so that they don't have to repeat the same basic questions and misunderstandings that the previous string of new members just hashed out. This would enrich the experience for all members and collectively encourage everyone to take their interest in the shared subject matter deeper, instead of burning out.

   This also, subsequently, feeds the AI tool, constantly improving the onboarding experience for new members as the community's collective knowledge grows.

   This only works if we are moving away from attention metrics and moving towards getting people what they are looking for quickly and effortlessly, affording them time to get back to living life and not getting sucked into an attention-hole on the internet every time they have a question.

4. Dispute Resolution

   This is a design space ripe for experimentation. One thing that is inevitable in a community is that people will have differing opinions about the subject matter.

   Sometimes this dispute cannot be resolved beyond agreeing to disagree. Not every question has a provable answer. But some disputes are rooted in misinformation that have to be constantly corrected. This is an area that has extremely ripe fruit to harvest for increasing the value proposition of a community tool. Anything that can help facilitate the resolution of commonly perpetrated disputes will reduce friction and tension in the community, resulting in a more positive and inclusive experience for everyone.

   This is the exact opposite of the Attention Economy. Web2 social media caters to these disputes because rage-bait is the single most effective way to capture people's attention for just long enough to inject an ad. Resolution is never the goal in this system.

   This is also an area prone to capture, however. I see this as a major flaw of both forums and modern social media communities: opinions tend to solidify around commonly repeated opinions of high value contributors. This can create an appeal to authority effect where the community whitewashes everyone's opinion to be uniform instead of embracing the diversity of reality.

   Experimentation in this area could find solutions where we don't algorithmically create confirmation bias feedback loops, but rather constantly push communities to be more open and inclusive to valid, but contrary opinions.

Make Communities Great Again

A common theme in the above 4 points is clear: Open Experimentation. There is no silver bullet, and no one-size fits all. So here are the 3 steps to bringing communities into the next age of social media:

1. Cut out the focus on attention space for ad revenue or engagement metrics, and maximize for serving the values and purposes of the communities: Knowledge Sharing.

2. Use that as the trojan horse to get people to accept blockchain rails. This creates an opportunity for people to start to learn about the features and functionality of blockchain that isn't just about casino games and number go up, without speculative incentives getting thrown in their face.

3. Incentivize app creation around making tools for content creators to do more than make financialized memes. Long-form discussion, educational content, how-to's, etc. are the foundation of a Community. The short-form, attention grabbing content can be built off that for growth and reach, but you can't skip to the end.

If all we do is cater to web3 communities that want to play casino games and pretend it's some kind of reasonable living, that's all we'll get. There are millions of communities out there that don't want anything to do with casino games that are just looking for a safe place to share their interests in a mutually beneficial way. This is proven by the fact that communities aren't migrating to web3 social platforms in hordes with the current focus on monetized speculative incentives.

This is our target market: content-driven communities trying to openly share knowledge.