Stephan pfp

Stephan

@stephancill

1338 Following
103557 Followers


Stephan pfp
Stephan
@stephancill
Send me a yo
27 replies
13 recasts
99 reactions

Stephan pfp
Stephan
@stephancill
Universal Basic Inheritance
0 reply
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
Tipping via /degensub 397 $DEGEN
0 reply
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
Tipping via /degensub 2252 $DEGEN
0 reply
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
Hell ye Faze up
0 reply
0 recast
1 reaction

Stephan pfp
Stephan
@stephancill
đź‘‘đź‘‘
1 reply
0 recast
4 reactions

Stephan pfp
Stephan
@stephancill
happy analysis friday everyone
0 reply
0 recast
4 reactions

Stephan pfp
Stephan
@stephancill
what prompted you to do this lmao
1 reply
0 recast
1 reaction

Stephan pfp
Stephan
@stephancill
afaik p0 of ethereum's client diversity is supposed to mitigate the risk of a client bug causing a consensus failure clients in ethereum serve a very different purpose to clients in farcaster
1 reply
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
wild idea. i kinda like it
0 reply
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
kind of! mbd would be one of many feed providers. and ideally feed providers would be open source or provide some proof that the feed was fairly ordered
0 reply
0 recast
0 reaction

Stephan pfp
Stephan
@stephancill
it just gets better
0 reply
0 recast
2 reactions

Stephan pfp
Stephan
@stephancill
occam's razor
0 reply
0 recast
1 reaction

Stephan pfp
Stephan
@stephancill
probably a good excuse to take a break from socials
0 reply
0 recast
1 reaction

Stephan pfp
Stephan
@stephancill
andrej karpathy says there have been 3 big iterations in how we build software 1. manual code 2. trained neural nets 3. instructing an LLM in natural language i love how he finds parallels that map so cleanly onto existing paradigms. highly recommend watching his recent talk https://www.youtube.com/watch?v=LCEmiRjPEtQ
0 reply
0 recast
2 reactions

Stephan pfp
Stephan
@stephancill
we should stop talking about decentralization (broad, hard to measure) and start talking about effective decentralization (is the achieved decentralization serving its intended purpose) example: replicating network data over thousands of nodes – high decentralization, but the data is accessed predominantly through a single node – low effective decentralization
1 reply
0 recast
5 reactions

Stephan pfp
Stephan
@stephancill
my assumptions might be wrong. time will tell
0 reply
0 recast
4 reactions

Stephan pfp
Stephan
@stephancill
getting increasingly bearish on the idea of client diversity as a means of effective decentralization key assumptions: 1. communities are very likely to use the same client 2. you're unlikely to migrate clients if you get nerfed that client, your ability to move clients doesn't actually help the situation unless your audience on that client also moves with you potential half-solution: separate the feed building layer from the client layer? not a bulletproof solution because no guarantee that you won't just get filtered at the client level anyway next level: encrypted feed building?
5 replies
0 recast
17 reactions

Stephan pfp
Stephan
@stephancill
Why would you willingly turn yourself into a bot? Automating replies like this adds nothing to the conversation and wastes everyone’s time Wish there was a way to mute all automated replies on X
3 replies
0 recast
15 reactions

Stephan pfp
Stephan
@stephancill
maybe naive of me, but i don't think it actually matters whether the actors building on top of a protocol are aligned. protocols shouldn't depend on this type of benevolence in farcaster's case i think the most relevant question is whether MM is building the protocol in a way that is fair and doesn't unfairly privilege their client over others and doesn't abuse their position as the developers of the protocol
0 reply
0 recast
0 reaction