I hope all 7994 of you are having a great week so far 🔥

If you're enjoying my writing, please share Terminally Onchain with your friends in Crypto and join the /toc channel on Farcaster 🤝

Sections Below

1. Channel Hail Mary

2. Kill the Hashtags

3. Chill if I add him to the chat?

Let's dive in.

On yesterday's Farcaster dev call, Dan discussed a new plan for channels as a hail mary to make them work in a meaningful way for Warpcast.

If it doesn't work, the team will wrap up the 1.5 year experiment and try exploring new avenues to drive growth to the network.

Before we dive in, let's quickly discuss channels in their current form factor.

In my Farcaster Channels post from March, I cover the story of how things got started in depth:

In June 2023, there was one week where a few events happened on the purple app that arguably transitioned the product into its next phase.

The first was the initial channel test on June 7th. Dan announced two use cases to start off with: FarCon 2023 & Purple.

Purple was an obvious choice as there was a ton of discussion going on regarding the DAO's mission, props, etc. And FarCon because it was set for that weekend in Boston!

The goal was to help clear out some of the content not fully relevant to the main feed (i.e. only a small group of casters were invested in the topic). One thing to note here is that anyone could still cast in and read the channels - the goal wasn't to gate content.

These early tests ended up being a success, which opened the floodgates for hundreds of new channels to spring up over the next year—first with some curation from the Merkle team, and then by opening them up permissionlessly to any users.

Today, there are over 3500 channels! Of course, the majority are small and inactive but it's clear that including the "subreddit" concept for Farcaster has made a huge impact on how people engage with content.

In fact, early on, channels brought a variety of new users to Warpcast because it allowed them to discuss non-crypto topics and explore niche interests.

But as Farcaster growth exploded in Q1 of this year because of frames and degen, the quality of content became difficult to manage.

There was an increasing need for moderation so channel owners could keep the ~vibes~ of their channels. Tools like Automod became essential to remove the spam/slop.

And now we're at the point where channels are clearly not pulling their weight considering how integral they are to the UX. There's confusion as to which channels users should post in. Large channels get slammed with non-sense. Channel discovery is tough. And channels have become a constant source of debate within the community.

So, what's the plan for channels to start pulling their weight?

Kill the Hashtags

One thing to understand about channels right now is that they come in two flavors.

The first is larger, general topics that resemble hashtags. Examples include /founders, /ai, /crypto, /books, etc.

And the second are what casters call "cozy corners", micro-communities that have a certain vibe, attract like-minded people, and are spaces that foster healthy conversation and debates. Examples include /latenightcrew, /ted, and /six.

Scrolling through "hashtag channels" feels like you're going through a part of twitter (i.e. tech twitter, AI twitter). And on the other hand, scrolling through "cozy channels" feels like you're reading through a public group chat.

By serving both use cases, channels have become a mess - it's not clear what purpose they're serving. So the Merkle team has decided to heavily lean into cozy corners. In the next update, hosts will manually add or invite people to channels just as you would in a group chat.

Why? Because the reality is that larger topics such as crypto and books are too broad! 90% of the casts in those channels can simply be regular posts to your home feed as one-off casts. For example, I've casted in books maybe 3-4 times total and I'd be totally fine if I just posted normally as I would have on Twitter.

In fact, in the last few months, I find myself personally casting my random thoughts or updates into the home feed anyways because I don't care to "place" them anywhere. I only consciously post into ~3 channels because they serve a specific purpose for me (i.e. /toc for Terminally Onchain or /latenightcrew). In those channels, I know the active members, it feels friendly for me, and there's a certain wavelength that all of us are on.

Let's take a look at a few examples of current hashtag channels that don't make sense in their current form. But when you think of them as umbrella topics for niche communities, then the purpose of the channel seems high utility.

It's worth noting that there can be a variety of channels under these umbrella topics that even serve the same purpose but cater to different types of people!

To pick on my own brand...not everyone on Farcaster will mesh with the general vibe of TOC but still want to discuss crypto topics in their own way. So they'll have a separate channel with a different style of discussion on crypto. Or with AI, there could be a channel for LLM devs but also another one for Replit fanatics that are obsessed with sharing tips & tricks.

And the awesome part is that channel discussion is public! So anyone can stop by, suss out the vibes, and interact if they resonate. If the host thinks that the user is providing meaningful engagement, then they can invite them and continue curating over time.

It's exactly like how you have a WhatsApp or Telegram chat and add people over time if you think:

1. They will provide to the conversation in a healthy way

2. Will mesh with the existing folks to a reasonable extent

One analogy that came to mind was student organizations on a university campus. When a student enters college, they share the campus and what it has to offer with everyone else there. But very quickly, they go to campus fairs to find clubs that interest them and make new friends with like-minded people. They end up finding 3-4 orgs that seem like a fit and use them to enhance their college experience.

Similarly, it's fair to expect new users who join Farcaster to initially see a huge campus full of all kinds of activity. But the goal will be to eventually find their own channels that feel like home in order to have memorable experiences and make new friends. As my fellow late night crew co-host @zinger likes to preach...friends > followers.

Can I add him to the chat?

With that being said, the examples of channels I gave above are easy to categorize and see how they will play out in the new model.

For my /toc channel, I'll just send an invitation to all my hypersubs. For /latenightcrew, we can send an invitation to anyone who has joined our weekly shows. And for something like a founders mastermind or bookclub, you're curating so that's straightforward.

But there are a whole set of channels that don't neatly fall into either hashtags or cozy corners. And that's where it'll be interesting to see how things play out.

For example, I host the /techarchive channel, a place to discuss the history of technology. Is it a cozy corner of people who are really passionate or is it a general hashtag? Many people find the topic interesting but I'm not really sure who I'm supposed to add. How do I know if you're knowledgeable about the topic and could be a "power caster"?

Maybe they'll reply to you but there's a good chance it'll get hidden with the "see more replies" button. And you can't always rely on the host to go through membership requests at a fast enough cadence.

Even location based channels or product channels become interesting case studies.

For /newyork, will it be just for people who live in the city and are actively hanging out with each other? But what if I'm visiting NYC and want recs for Thai food...is that good enough to get added to the channel? You probably won't be added if you're just a tourist who will post the millionth picture of the skyline at sunset time.

And for the /paragraph channel, will it just be the core Paragraph team? Or will they add their most active creators? Or any creator who even has a paragraph? What if someone wants to give one-off feedback on the product?

Some answers that come to mind for me:

1. Not clear who the target audience is? Then niche down another layer. With /new-york, it might be that there needs to be a /live-in-nyc and a /nyc-recs channel.

2. Channel Admins - after I add the first 5-10 people in the /techarchive channel, the hope is that they'll see content that seems relevant in their feeds, quote cast into the channel, and then I can invite the new caster. Later down the road, there can be an option to make some trusted channel members admins that can add people as well. Similar to how there's a WhatsApp chat admins that have some privileges but aren't owners.

3. Onchain proof of work - with the /paragraph channel, the team can check to see if the user's wallet address has created or collected a post on the platform. Or the /nouns channel can check if you own a Nouns NFT and have voted on an onchain proposal.

Of course, these things end up working out in mysterious ways so it'll be interesting to see how different community owners decide to curate their channel.

In the short term, there will definitely be chaos, pushback, etc. Many of us who got used to posting in channels for the majority of casts will need to retrain posting behavior to just cast in the home feed. Better discovery of content in home feed will be a never ending process. And most importantly, perfecting the flow for new users will be crucial to make cozy corners actually work.

Overall, I'm glad that there's changes happening to the current channel model. I think the first few versions served their initial purpose of promoting diversity of content. But now it's time to lean into what's working and focus on keeping content as high quality as possible.

That's all for today's post, I hope all of you have a great weekend!

If you have any thoughts on this topic, I'd love to hear them in the TOC channel public-read group chat.