Farcaster hosted their first developer day this week, and it was a blast!

I connected + reconnected with internet friends, tried out new products, and demo'd my own.

The community was excited as ever, but also equally pragmatic about the journey ahead.

I could not be more bullish on the future of Farcaster.

We started the day with a discussion moderated by Dan and Varun (co-founders of Farcaster).

The main topics were:

• Growth

  • Retention

  • Channels

  • Frames

• DC Roadmap

• Reducing sign up costs

• Hubs - stability + scalability

• Other

We talked about 4 main ways to drive growth:

1. Retention - without it, you can acquire as many users as you want and still not grow

2. Channels - by creating "cozy corners", we can improve + increase community building on FC

3. Frames - they're the best distribution channel for new crypto apps today

4. Hubs - they have to be stable to withstand the next phase of network growth

• Channels, Frames, and Mini-Apps are the keys to retaining Farcaster's current set of users

  • Channels help users connect with like-minded people

  • Frames help developers get distribution for their products

  • Mini-Apps give users more ways to use the network without leaving their feed

The focus for channels is to help people curate their own communities.

The thinking is that it's easier to moderate people than it is to moderate content.

To roll this out, the Merkle team is focused on:

1. Membership: Starting next ~Thursday, you'll have to be a member of a channel to cast in it

2. UX refresh: Users will notice this change to channels starting next week

3. APIs: Developers can start building with the API today (to add / remove channel members)

After a successful rollout, the focus will shift to:

1. Allowing new users to onboard directly into a channel

2. Decentralizing the channel primitive into the Farcaster protocol

Frames are the best way for developers on Farcaster to get distribution for their products.

They can be simple (ie. allowing people to take action in an app that lives in feed) or complex (ie. loading a full web page / mini app in feed).

Unlike typical OpenGraph images, clicks on a Frame contain an authenticated FID, giving developers more info about who is interacting with their Frame.

The Warpcast team has also built an Explore page to help current users find Frames they've interacted with in the past and help new users discover Frames they might want to interact with in the future.

Mini apps were a hot topic at dev day.

If Frames are used as the "advertisement" for your app, Mini Apps let users try out your app — all without leaving their feed.

The main value I see here is how few clicks it takes to get a user into your app's experience.

On Twitter, a user would need to...

1. Click a link in feed

2. Open a new tab

3. Sign up for the application

4. Try to use it

On Farcaster, a user would only need to click a link in their feed.

By cutting out steps, Mini Apps drastically decrease time to value for your product.

This is a game changer for acquiring and retaining users on Farcaster.

Direct Cast usage has been stable over the past few months, but programmatic DCs are growing.

In the short term, the goal is to improve programmatic DC functionality + performance.

In the longer term (2025), the Merkle team wants to increase DC interoperability, allowing developers to read/write DCs in their own apps and reducing the dependency on Warpcast for direct messaging.

One interesting fact: if a user sends/receives a direct cast within their first 7 days on Farcaster, they retain at a higher rate than a user who did not send/receive any direct casts. This insight was shared in passing, but I think it's valuable for those looking to build on Farcaster, and those looking to help grow the network more broadly.

I've seen quite a few complaints about paying to sign up for Farcaster.

But if you look deeper, the cost has decreased with time — and will continue to do so.

By early 2025, the goal is to get sign up costs down to $1.

The main blocker to doing so is spam — the lower the sign up cost, the more spam on the network.

Spam has and will continue to be a problem as the network grows, and it remains top of mind for Merkle and indie devs alike.

TL;DR - syncing between Hubs is improving, but it's rough around the edges.

To withstand the next step change in network growth, we'll need a redesign (see Snapchain).

Prototyping has begun, and if you want to contribute, the highest leverage actions are:

1. Poke holes in the Snapchain design (discussion here)

2. Help write code for the implementation

To close out the morning's presentation, we quickly discussed a few other points

1. More data will be coming to the protocol soon (like location - finally!)

2. Sign in with Farcaster will add web + multi-client functionality (expanding beyond mobile-only)

3. Neynar + OpenRank are working on spam datasets for developers who want cleaner feeds

After lunch, many developers demo'd their products — and wow, was I impressed!

So much alpha was leaked that afternoon. Here's what I remember:

• Christopher + Erica demo'd Uno (a new image-based, "algotorial" client)

• Will + Raihan demo'd an unreleased product from Syndicate

• Manan demo'd Neynar webhooks

• Mikey demo'd MoshiCam

• Michael demo'd Buoy

• Matthew

This afternoon made me feel incredibly energized.

The Farcaster community isn't just a bunch of cracked developers — it's a passionate group that cares deeply about supporting each other's ambitions.

As a new developer, that meant the world to me.

Huge shout out to Ted for curating an amazing group, and for making this special day happen.

Also thanks to Dan and Varun, without whom this network wouldn't exist.

And final thanks to everyone I met and spent time with — you are all one of a kind.

After a long night's sleep, and some time reflecting today, my big takeaway from all of this is the power of community.

Farcaster is interesting on the surface as a place to permissionlessly develop social apps.

But go a layer deeper, and you'll see one of the most talent dense ecosystems in crypto.

Beyond sheer talent, Farcaster's community is kind, welcoming, and beyond supportive.

Without the people who make it what it is, Farcaster would not be where it is today.

And for that, I could not be more grateful.

Farcaster hosted their first developer day this week, and it was a blast!

I connected + reconnected with internet friends, tried out new products, and demo'd my own.

The community was excited as ever, but also equally pragmatic about the journey ahead.

I could not be more bullish on the future of Farcaster.

We started the day with a discussion moderated by Dan and Varun (co-founders of Farcaster).

The main topics were:

• Growth

  • Retention

  • Channels

  • Frames

• DC Roadmap

• Reducing sign up costs

• Hubs - stability + scalability

• Other

We talked about 4 main ways to drive growth:

1. Retention - without it, you can acquire as many users as you want and still not grow

2. Channels - by creating "cozy corners", we can improve + increase community building on FC

3. Frames - they're the best distribution channel for new crypto apps today

4. Hubs - they have to be stable to withstand the next phase of network growth

• Channels, Frames, and Mini-Apps are the keys to retaining Farcaster's current set of users

  • Channels help users connect with like-minded people

  • Frames help developers get distribution for their products

  • Mini-Apps give users more ways to use the network without leaving their feed

The focus for channels is to help people curate their own communities.

The thinking is that it's easier to moderate people than it is to moderate content.

To roll this out, the Merkle team is focused on:

1. Membership: Starting next ~Thursday, you'll have to be a member of a channel to cast in it

2. UX refresh: Users will notice this change to channels starting next week

3. APIs: Developers can start building with the API today (to add / remove channel members)

After a successful rollout, the focus will shift to:

1. Allowing new users to onboard directly into a channel

2. Decentralizing the channel primitive into the Farcaster protocol

Frames are the best way for developers on Farcaster to get distribution for their products.

They can be simple (ie. allowing people to take action in an app that lives in feed) or complex (ie. loading a full web page / mini app in feed).

Unlike typical OpenGraph images, clicks on a Frame contain an authenticated FID, giving developers more info about who is interacting with their Frame.

The Warpcast team has also built an Explore page to help current users find Frames they've interacted with in the past and help new users discover Frames they might want to interact with in the future.

Mini apps were a hot topic at dev day.

If Frames are used as the "advertisement" for your app, Mini Apps let users try out your app — all without leaving their feed.

The main value I see here is how few clicks it takes to get a user into your app's experience.

On Twitter, a user would need to...

1. Click a link in feed

2. Open a new tab

3. Sign up for the application

4. Try to use it

On Farcaster, a user would only need to click a link in their feed.

By cutting out steps, Mini Apps drastically decrease time to value for your product.

This is a game changer for acquiring and retaining users on Farcaster.

Direct Cast usage has been stable over the past few months, but programmatic DCs are growing.

In the short term, the goal is to improve programmatic DC functionality + performance.

In the longer term (2025), the Merkle team wants to increase DC interoperability, allowing developers to read/write DCs in their own apps and reducing the dependency on Warpcast for direct messaging.

One interesting fact: if a user sends/receives a direct cast within their first 7 days on Farcaster, they retain at a higher rate than a user who did not send/receive any direct casts. This insight was shared in passing, but I think it's valuable for those looking to build on Farcaster, and those looking to help grow the network more broadly.

I've seen quite a few complaints about paying to sign up for Farcaster.

But if you look deeper, the cost has decreased with time — and will continue to do so.

By early 2025, the goal is to get sign up costs down to $1.

The main blocker to doing so is spam — the lower the sign up cost, the more spam on the network.

Spam has and will continue to be a problem as the network grows, and it remains top of mind for Merkle and indie devs alike.

TL;DR - syncing between Hubs is improving, but it's rough around the edges.

To withstand the next step change in network growth, we'll need a redesign (see Snapchain).

Prototyping has begun, and if you want to contribute, the highest leverage actions are:

1. Poke holes in the Snapchain design (discussion here)

2. Help write code for the implementation

To close out the morning's presentation, we quickly discussed a few other points

1. More data will be coming to the protocol soon (like location - finally!)

2. Sign in with Farcaster will add web + multi-client functionality (expanding beyond mobile-only)

3. Neynar + OpenRank are working on spam datasets for developers who want cleaner feeds

After lunch, many developers demo'd their products — and wow, was I impressed!

So much alpha was leaked that afternoon. Here's what I remember:

• Christopher + Erica demo'd Uno (a new image-based, "algotorial" client)

• Will + Raihan demo'd an unreleased product from Syndicate

• Manan demo'd Neynar webhooks

• Mikey demo'd MoshiCam

• Michael demo'd Buoy

• Matthew

This afternoon made me feel incredibly energized.

The Farcaster community isn't just a bunch of cracked developers — it's a passionate group that cares deeply about supporting each other's ambitions.

As a new developer, that meant the world to me.

Huge shout out to Ted for curating an amazing group, and for making this special day happen.

Also thanks to Dan and Varun, without whom this network wouldn't exist.

And final thanks to everyone I met and spent time with — you are all one of a kind.

After a long night's sleep, and some time reflecting today, my big takeaway from all of this is the power of community.

Farcaster is interesting on the surface as a place to permissionlessly develop social apps.

But go a layer deeper, and you'll see one of the most talent dense ecosystems in crypto.

Beyond sheer talent, Farcaster's community is kind, welcoming, and beyond supportive.

Without the people who make it what it is, Farcaster would not be where it is today.

And for that, I could not be more grateful.