Varun Srinivasan pfp
Varun Srinivasan
@v
An interesting Farcaster idea: what if apps paid for storage, not users? - Apps buy a chunk of bytes. - Free to allocate to casts, reactions, follows in any way. - Free to assign to users in any way (fixed rate, free) - Lower cost by avoiding "unused storage" h/t to @vrypan.eth @deodad and @sds
23 replies
73 recasts
428 reactions
Varun Srinivasan pfp
Varun Srinivasan
@v
Main benefits are that: 1. Cost of onboarding new users is lowered. Apps can front users and kick them out if they are spammy later. 2. Storage utilization is more efficient. Lots of users overpay and don't use bytes, apps could get better economies of scale.
2 replies
2 recasts
126 reactions
vrypan |--o--| pfp
vrypan |--o--|
@vrypan.eth
Interesting! - How does pruning work in this model? Per app or per user? - Can a user easily move their data between apps? (yes, resigning and resubmitting, but are there any other limits they may hit in this case such as throttling?) - How could the ordering architecture be affected by this change? What if apps did not buy storage, but tipped sequencers? Many models, not auctions necessarily. For example a flat fee, but pick the sequencer.
1 reply
0 recast
1 reaction
vrypan |--o--| pfp
vrypan |--o--|
@vrypan.eth
Let's have a placeholder for this: https://github.com/farcasterxyz/protocol/discussions/200
1 reply
0 recast
0 reaction
Johnson pfp
Johnson
@johnson
Apps would have to be able to pay only for storage of casts users posts from the app rather than other apps right? So they don't end up paying for users who are active but not just from their app.
1 reply
0 recast
3 reactions
jcdenton.cast pfp
jcdenton.cast
@jc
Will users still be able to pay for themselves under this model? e.g. “be their own app” Also how are FIDs created - would it be free since apps pay storage now? e.g. can I onboard 100,000 FIDs with 1 unit? It’s an interesting idea!
1 reply
0 recast
2 reactions
Liang @ degencast.wtf 🎩 pfp
Liang @ degencast.wtf 🎩
@degencast.eth
if apps pay for storage, do they get yield from sharing the data to the protocol?
1 reply
0 recast
3 reactions
Jawa pfp
Jawa
@jawa
I love the idea of apps paying. But what would really be great is if we had a clear idea of what we are paying for and what it means. Does paying for our own data storage mean that FC won’t sell our data to train LLMs? That would be a huge selling point.
1 reply
0 recast
0 reaction
Stephan pfp
Stephan
@stephancill
This model feels wrong Farcaster users don’t necessarily have a relationship with a single app. What happens when the user decides to stop using the app that pays for their storage? Feels like there’s an incentive misalignment
1 reply
0 recast
0 reaction
Greg pfp
Greg
@greg
Initial take is that this feels like a complicated solution to something that doesn’t seem like a huge problem rn. Doesn’t Warpcast sponsor signup fees for many users anyways?
1 reply
0 recast
0 reaction
depatchedmode pfp
depatchedmode
@depatchedmode
Feels right.
0 reply
0 recast
1 reaction
chris 🤘🏻 pfp
chris 🤘🏻
@ckurdziel.eth
I think this makes a lot of sense - apps could implement tiers and/or pass optionality on to users as they see fit if they want to store more. Probably a win for user growth overall
0 reply
0 recast
1 reaction
Linda Xie pfp
Linda Xie
@linda
Pretty open to this! Would suggest that app developers can get some small grants early on to cover the storage costs in addition to the operational overhead
1 reply
0 recast
5 reactions
​woj pfp
​woj
@woj.eth
100% agree it should be this way
0 reply
0 recast
1 reaction
William Saar pfp
William Saar
@saarw.eth
Makes a lot of sense! Wrote a hackathon project that used Solana's functionality to let third parties pay for on-chain transactions so apps could pay for users to interact with a zk verification oracle and get a token. Idea was the payment could come from an app's anti-spam budget https://github.com/saarw/nymcred
1 reply
0 recast
1 reaction
Liang @ degencast.wtf 🎩 pfp
Liang @ degencast.wtf 🎩
@degencast.eth
100% love to involve apps into the value chain tho
0 reply
0 recast
1 reaction
christopher pfp
christopher
@christopher
i can buy this. it's how apps operate today in centralized data models. servicing the storage is just part of our OPEX.
0 reply
0 recast
1 reaction
Sinaver pfp
Sinaver
@sinaver.eth
does it mean we're finally moving towards more granual storage approach? Can user buy a chunk too or it will be available only for apps? Can an app be a market itself in that case: buys cheaper, sells cheaper to the user (like with gift cards)? I think overall I like the idea because this gives an answer for people saying you need to pay for storage to try farcaster: just use another client which gives initial storage for you for free!
0 reply
0 recast
0 reaction
Britt Kim pfp
Britt Kim
@brittkim.eth
Is it possible that apps ruin the FC experience due to noob storage management (not providing sufficient head room, real time storage increases, etc.) and this leaves the user powerless and frustrated? Would the user really be able to distinguish between bad app and bad protocol here? Maybe worth building out a solid storage library for apps to prevent common mistakes. I say this because roll-your-own FC login hasn’t been the best experience, so my confidence in apps isn’t high.
0 reply
0 recast
0 reaction
Liang @ degencast.wtf 🎩 pfp
Liang @ degencast.wtf 🎩
@degencast.eth
I like this framework https://x.com/0xCygaar/status/1847354267436535874?t=voPBSU9zqAwhQOqtURxV6g&s=19
0 reply
0 recast
0 reaction