Can others volunteer inputs to improve the anonymity set?
Yesterday Bull Bitcoin (npub1t289s8ck5qfwynf2vsq49t2kypvvkpj7rhegayrur0ag9s2sezaqgunkzs) became the first mobile wallet to support async payjoin (BIP-77). Congratulations to nostr:npub1yevrvtp3xl42sq06usztudhleq8pdfsugw5frgaqg6lvfdewfx9q6zqrkl and the payjoin team!
This is means you can now use payjoin without running your own server. Want to learn how it works? Jump to page 7 of the payjoin zine.
To read it online or print copies of your own, visit https://satsie.dev/zines/payjoin.html
Discussion
I think that might be what they are working on for the next version of payjoin 🙂
That's good, a 2 of 2 anonymity set isn't great, but better then no Payjoin at all!
“Anonymity set” as understood in legacy CoinJoin parlance is a bogus half-baked idea that breaks down as soon as transfers get made. this iteration of payjoin doesn’t address that. It breaks Common Input Heuristic.
Doesn't legacy coinjoin also break common input heuristic since all inputs are the same?
Why is Legacy "bogus" and v2 not?
Not really, since coinjoins have an overt signature that can be identified as a coinjoin. Payjoin looks identical to many other types of typical transactions that any wallet would mak before payjoin existed.
The Anonymity Set metric claimed to measure coinjoins isn’t robust since your peers in the join can deteriorate it by revealing themselves. The way it has been deployed makes no guarantees about the costs an attacker would need to pay to violate your privacy. Not to say anonymity sets aren’t real, they are, but the way that word has been applied has been deceiving.
This exchange was very informative 🙏