Please link the unwound joinmarket transaction. Last I checked there was still a 100s btc bounty for unwinding some dudes stolen btc.
Unless you have a real life example of someone getting footgunned by joinmarket your points are moot.
https://www.reddit.com/r/Bitcoin/comments/69duq9/50_bounty_for_anybody_recovering_445_btc_stolen/
they absolutely are not moot.
anonset matters.
Sybil resistance matters.
explain to me how a user can measure the privacy gained by a JM coinjoin.
also
https://bitcointalk.org/index.php?topic=1609980.msg16168594#msg16168594
also I think Eego got into this on Liveras pod here
but may be misremembering
You're a little retarded. Did you even read the first reply to your 8 year old link? It's well documented that one coinjoin is insuffiecient for privacy (https://github.com/JoinMarket-Org/joinmarket-clientserver/blob/master/docs/USAGE.md#try-out-a-coinjoin-using-sendpaymentpy).
>explain to me how a user can measure the privacy gained by a JM coinjoin
Someone stole 445 btc and used joinmarket to mix it. The theft victim offered half the btc if anyone could unmix it. No one could.
oh so what you MEANT to say was
"CJs can be unwound and its well documented,
but if you do enough of them you PROBABLY have reasonable privacy guarantees"
also
one bounty doesnt prove shit.
we need good privacy guarantees at scale.
not this "well then tell me who owns this UTXO" horseshit
Are you aware of any mixing protocol anywhere that is sufficient with just one round?
the whole point of a proper Zerolink implementation is there's no way to know which input matches which output.
even after a single mix.
as I understand it, Wasabi didnt end up implementing it completely, but Whirlpool did (with modifications).
this is not to shit on JM, just that people know the tradeoffs.
you can get reasonable privacy with JM.
probably.
its all a probability game.
