Varun Srinivasan on Warpcast

Varun Srinivasan pfp

Varun Srinivasan

We're considering ending support for v1 Frames in 3 months. Why? - It has little usage; most devs moved to miniapps - It simplifies the protocol for client builders + new devs. - It simplifies the UX for users. Only one type of "app" in the feed. Over 90% of devs we polled support this decision.

27 replies

14 recasts

110 reactions

Varun Srinivasan pfp

Varun Srinivasan

v2 Frames (now Miniapps) launched in November 2024. The eventual goal was to replace v1 Frames, but we supported both to make the transition easy for developers. Five months later, most developers have moved onto miniapps and are supportive of simplification. We plan to give developers another 3 months notice before ending v1 Frames support.

1 reply

0 recast

29 reactions

phil pfp

I’m fine with the change. At some point, please consider having better support for the cast action to miniapp pipeline.

1 reply

0 recast

3 reactions

eggman 🔵 pfp

v1 was a fun spec to write for, never wrote up that sorta render loop before. Pretty proud of it too, the caching system made it feel super speedy even when the entire protocol were hammering out waifus. But yeah, it's pretty clear the functionality of v2 (and no actual NEED for a custom render/caching system) has shown how unwieldy v1 was in retrospect. Am going to miss the in-feed embeds though. Genuinely did like those, but yeah, not very compatible with v2 unless you actively load in and launch every single miniapp before rendering the home screen. i.e: set devices on fire. Good show, long live Frames 🫡

0 reply

3 recasts

13 reactions

niftytime pfp

support ending v1 frames most v1 frames are like glorified hyperlinks anyways (which for folks less familiar with the distinction between v1 frames and mini apps can lead to some damaging disappointment as most new users will be expecting mini app level experiences imo)

0 reply

0 recast

2 reactions

Marcelo Terça-Nada 💎🎩✨ pfp

Marcelo Terça-Nada 💎🎩✨

Varun and @dwr.eth I completely understand this decision to end support for v1 frames. One negative impact of this will be that we will no longer have working frames from platforms like @fxhash @objktcom @superrare (or maybe this is a good time to reach out to them and invite them to return to Farcaster).

0 reply

0 recast

1 reaction

Colin Johnson 💭 pfp

Colin Johnson 💭

Good to kill V1 frames, but I would strongly advocate for one tap actions for in-feed cards. Polling products are a simple type of interaction that require minimal context and that provide great entertainment/insight when consumed at scale. Putting poll answers inside of a mini app effectively eliminates that category of interaction, or reduces engagement by ~70%. Outside of polling, my expectation is that one tap transactions in feed would increase the velocity of the Farcaster economy more than any other feature (maybe 20-40% above baseline). Our ideal solution would be for anyone who’s added the frame to get a personalized view that includes up to 4 single tap options if the developer chooses, else falls back to the single launch button. This way timelines stay clean for users who haven’t shown interest in the app, but retain some of that “meet users where they are” magic.

2 replies

3 recasts

5 reactions

Mark Carey 🎩🫂 pfp

Mark Carey 🎩🫂

I continue to strongly believe that decommissioning v1 frames would be a mistake, that doing so would hurt the Farcaster dev ecosystem Why? - v1 frames enabled users to take simple app actions without leaving the feed, without needing to open an app. There are more steps/barriers to participation to engage with a mini app (one click to open, then most likely met with dialog to add frame and/or sign in, then do the desired action) - v1 frames have a lower barrier to entry for devs, since they are effectively headless apps (devs do not need to code client-side html/css/js). Proposal: - replace v1 frames with a similar in-feed mini app format - effectively, these would look like mini app embeds -- and thus consistent/familiar to users -- except the button(s) act more like v1 actions instead of opening mini app - give them access to notifications - enable them to be "added" just like full mini-apps - maybe limit them to 1 or 2 "in-feed" actions buttons (if devs need more, then maybe a full mini app is better)

1 reply

1 recast

3 reactions

borodutch pfp

@warpcastadmin.eth

please allow us to tailor-make mini-app initial previews by passing ?fid=123 to the request 🫠 then i'll be fully on board with deprecating v1 the feed will get way more vibrant because miniapps will change based on who's looking!

2 replies

0 recast

3 reactions

dylan pfp

all in on miniapps 🫡

0 reply

0 recast

3 reactions

Pichi 🟪 🍡🌸 pfp

Pichi 🟪 🍡🌸

@undefined making sure you see this thread.

1 reply

0 recast

3 reactions

Matthew Fox 🌐 pfp

Matthew Fox 🌐

0 reply

0 recast

3 reactions

Darryl Yeo 🛠️ pfp

Darryl Yeo 🛠️

RIP 🫡 https://warpcast.com/darrylyeo/0x72531fb3

0 reply

0 recast

2 reactions

Hoot 🎩 pfp

Would like to see more in-feed customization for mini app embeds before the end of life of v1. eg more than one button, or in the future even slight interactive elements ala @mod or Googles AMP?

0 reply

0 recast

0 reaction

matthew chaim pfp

agree with this. v1 feels very v1. time to move on and go all in on the Mini App

0 reply

0 recast

0 reaction

jacopo.eth pfp

agreed, all-in on mini apps

0 reply

0 recast

0 reaction

Koolkheart pfp

@koolkheart.eth

We need to do a burial ceremony for frames

0 reply

0 recast

0 reaction

raulonastool.eth 🎩 🏰 pfp

raulonastool.eth 🎩 🏰

I really hope you don't. I like v1 frames. They felt native to the feed. like little windows you could tap without leaving the moment of being in a social app. V2 is powerful, but it pulls you out of the experience. We need both. Windows and portals. Unless you can make a v3 Frame that does both, I say lets keep them.

0 reply

0 recast

2 reactions

Plants pfp

Actions need love. Those are valuable and unique to this space

0 reply

0 recast

1 reaction

Andrei O. pfp

Frame v1 was suboptimal because it used images as transport, and all the power of the Mini-App is backed by Iframe which is a much better transport, plus it allows client-only code and any kind of UI. That being said, when you open an Iframe, the potential for abuse is smaller since the Iframe at least starts from user interaction. But is trivial for a bad actor to kill the client by running malicious resource-consuming code like an event bomb to kill the main thread and crash everything, so it means that Iframes directly in the feed is a big no, no, so that's the reality people will have to settle with these MiniApps that have huge potential only after you open/first interacted with them, is a good-enough primitive.

0 reply

0 recast

1 reaction