Cover photo

/memes report #1

In order to FO you must first FA: background, insights, and next steps for the /memes channel subscription experiment.

Channel Stats

  • Revenue: 0.78 ETH

  • Followers: 235,234

  • Casts: 16.8k

  • Recasts: 50.8k

  • Replies: 104.8k

  • Likes: 451.4k

  • Churn: 15%

Data from CastSense and Hypersub.


On May 14, Dan Romero shared a proposal for decentralizing Farcaster channels:

I've been bullish on channels as a meaningful differentiator for Farcaster, in particular because they can be owned and independently managed using powerful third-party tools and onchain primitives like NFTs and ERC-20 tokens.

This opens up a huge design space that, once explored, could yield new uses cases and novel interactions that materially increase network growth.

Consider the /degen channel, which was reluctantly launched as a response to vocal users and ultimately became one of the primary drivers of Farcaster's growth.

/degen bootstrapped its own community—culture, currency, and economy—all by itself, by using simple onchain primitives, and creating new ones like reply tipping along the way. Now that channels are decentralizing and new building blocks are being shipped every day, it's inevitable that more and more channels will begin to develop their own cultures and economies.

It could be that /degen was the beginning of something much bigger, and provides a repeatable playbook for other channels to use and benefit from.


To better understand decentralized channels, and to get a feel for how different tools can be combined to customize their experience, I created /hypermod.

Hypermod was an experimental channel that used Hypersub and Automod to display casts by channel subscribers, and hide casts by non-subscribers.

Here's a video of how I set it up, and how it worked:

A week after Dan's initial cast, he announced Phase 1 would roll out in two days:

Since the Hypermod experiment worked well, I thought why not try this with /memes?

So the next day I deployed the Hypersub, airdropped it to active casters, and documented my process for setting up the channel.


The idea was simple:

  1. Use Hypersub to charge users a monthly fee for their casts to appear in the Main feed.

  2. Use Automod to check if a user is subscribed and either display their cast in the Main or Recent (spam) feed.

  3. Use a Splits contract as the recipient address for the Hypersub revenue, and divide the funds 31/69: 31% to me personally, 69% to a tipping pool.

  4. Use Paybot to tip top casts from the tipping pool.

The annual price of the subscription is .0042069 ETH (~$16) and the amount I began tipping was also .0042069 ETH (~$16). Which means that with just one banger cast, the entire annual subscription would be paid for. After a second banger, users could start earning.

The thesis is that as a result, both channel quality and engagement will increase.

My grand vision for the channel is that it becomes the best place on the internet for finding the dankest memes.


Thinking about it more, I realized I could further share the workload, and the upside.

What if instead of just me moderating, earning, and tipping, I divided up the pie with others? I reached out to two active casters in the /memes channel with the following proposal:

They agreed and I updated the splits contract as follows:

Here's a video of how I set it up, and how it worked:


More and more channel owners are beginning to experiment with this approach:

The more I explain how the various building blocks can be put together, and the more I help others configure their channels, the more I realize I need to simplify.

The current /memes channel configuration is powerful, but difficult to setup and explain.

Next Steps

Moving forward I will be reverting to the previous configuration, removing the two moderators I added, and updating the splits contract to a 4/96 split: 4% to me personally, 96% to the tipping pool.

This is by no means the end state. The plan is to keep it simple for the next month or so, then progressively add mods back as needed.

I also plan to experiment with and other emerging forms of sharing value onchain.

If you have any questions or suggestions for what can be improved, please leave a comment below or send me a message on Farcaster.



Fabric logo
Subscribe to Fabric and never miss a post.
  • Loading comments...