Stephan pfp

Stephan

@stephancill

817 Following
91151 Followers


Stephan pfp
Stephan
@stephancill
openai cofounder says he expects to see AGI in 2-3 years lol good interview https://youtu.be/Wo95ob_s_NI
3 replies
1 recast
4 reactions

Stephan pfp
Stephan
@stephancill
Sounds like something that needs to exist. I haven’t built saying in this space but I know df is looking to turn farcaster.id into a router of sorts which opens links in a preferred client Would love to learn more about the architecture you’re thinking about to achieve this
1 reply
0 recast
1 reaction

🎩.artlu.eth👃- 呂Kunst pfp
🎩.artlu.eth👃- 呂Kunst
@artlu
trying smth - - 4 clicks to get from Warpcast to Supercast - can become 2 clicks when some training wheels get taken off I will use this simplistic action to build portable, private, composable FC-ecosystem-wide onchain user settings (e.g., specify preferred clients) https://fc-clients-cast-action.artlu.xyz
19 replies
8 recasts
44 reactions

Stephan pfp
Stephan
@stephancill
Cc @df /farcaster-id
1 reply
0 recast
3 reactions

Stephan pfp
Stephan
@stephancill
i see this as fair use
0 reply
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
evergreen tweet
4 replies
1 recast
6 reactions

Stephan pfp
Stephan
@stephancill
how does eth capture value from memecoins on L2 🫠 if anything, people are dumping eth to ape into memecoins
2 replies
0 recast
1 reaction

Stephan pfp
Stephan
@stephancill
BLEU
5 replies
4 recasts
16 reactions

Stephan pfp
Stephan
@stephancill
Holy shit lol
0 reply
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
it doesn't matter if it's 100 commands or 1 command, when the state is multiple terabytes it's going to be hard for people to download & store. this is going to be a problem for all decentralized social networks evetually, which is why it's important to be extremely conservative with what you put in the network state
0 reply
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
bluesky also gets this right
1 reply
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
let's not conflate "onchain" with "open & decentralized" blockchains have a lot of overhead to guarantee the exact ordering and atomicity of transactions which are not necessary for social and result in additional costs for end users this is what lens gets wrong vs farcaster imo
2 replies
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
i was also surprised when i saw sama's tweet. i thought some people had access already so assumed it was a phased rollout and have been checking my chatgpt every now and again to see if i have access lol
0 reply
0 recast
2 reactions

Stephan pfp
Stephan
@stephancill
the whole point of SNARK proofs is that they can be validated cheaply - the problem is that it is very expensive to generate them
1 reply
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
operational costs *after migrating
1 reply
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
my concern is that the benefits using of a blockchain diminish when it becomes very hard to sync and run a node wrt cost - posting and verifying validity proofs is cheap, but computing them is not. i would be curious to hear what you expect the total operational costs of migrating lens to a validium will be
1 reply
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
okay cool, i like that!
0 reply
0 recast
1 reaction

Stephan pfp
Stephan
@stephancill
nice! what about the 1 character usernames 😆
1 reply
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
fyi the new chatgpt voice feature isn't live yet. any voice feature you've been using has been there for months already easy way to check if it's the new version is to ask it to sing something https://x.com/sama/status/1790817315069771959
2 replies
3 recasts
29 reactions

Stephan pfp
Stephan
@stephancill
can you elaborate on what you mean by the modularity of the architecture? what are devs' options when it comes to putting things onchain? my understanding is that posts, reactions, profile data, and follows are the primitives that always go onchain. is that right? thank you!
1 reply
0 recast
1 reaction