I may be wrong on some of these, but as I understand it thus far...
- Cashu operators are custodial.
- But there's no formal account registration process. Similar to some LNBits
- The amounts are blinded so the operator doesn't know who/how much
- Some mints integrate to other mints and lightning for swaps
- The above can be leveraged to anonymously send/redeem value through the network in a privacy preserving way
So if cash is successful at making this work, then coinjoins are kind of a thing of the past.
I know there are some small differences and benefits to both, but it seems cashuâs trade offs are significantly better. It doesnât cost much, itâs much more private and itâs almost instant.
Is my argument mostly correct?
Collaborative coinjoins are still appropriate for layer 1 transactions
But why pay 5% fee, wait for transaction times, etc when you could just cashu funds to a different wallet you control and no one is any wiser?
Because collaborative coinjoins don't have counterparty risk which you are taking on whenever you use cashu? Decisions and risk tradeoffs based on size.
Coinjoins are still very inconvenient and most people donât know how to do them successfully without fucking up their privacy. All it takes is 1 mistake. Even pros mess it up.
Also.. 5% fee seems awfully high. For whirlpool this would only be that high if you were mixing only 1 output worth. It's more effectivr to enter a mix with more, producing more outputs to reduce the effective fee and then stay remixing for free.
Can you explain a bit more? It sounds promising but I donât know what you mean quite yet.
Whirlpool has a few different pool sizes.
.5 BTC, .05 BTC, .01 BTC and .001 BTC
Let's say you have .2345 BTC
With this amount you can't use the .5 BTC pool, but could enter any of the others.
If entering the .05 BTC pool, the fee would be .0025, and you end up with 4 outputs being mixed and change. Effective fee of 1.25% per output (.0025 / .20)
If entering the .01 BTC pool, the fee would be .0005 BTC, and you'd get 23 outputs. The effective fee per output is then about .2173% per output (.0005 / .23)
There are maximums on the amount you can enter a pool with for Tx0. I think the .001 BTC pool is limited to at most 25 UTXOs (and the others up to 75?) so if that's true then with .2345 BTC, you'd only be able to enter that .001 BTC pool with .025 BTC with a pool entry fee of .00005. This makes for an average of .2% (.00005 / .025) per output
Ahh yes, makes perfect sense now. Thank you and zapped!
Coinjoins are for on-chain privacy. Cashu is a token that gets exchanged off-chain
Right, but I donât care if itâs on chain or not. The idea is if no one knows who has it after itâs sent thru cashu then why coinjoin?
Do you mean something like: on-chain -> lightning -> cashu -> different lightning wallet -> different on-chain wallet?
Yes indeed. Much cheaper and easier to do then a coinjoin.
I guess the difference is that on-chain you still just had one transaction. Nobody knows if itâs you who received it or somebody else, but neither does anyone know that when I just do a normal transaction. But there is only one possible history. With a coinjoin the link to the history of the UTXO is broken, because a hand full of people does a transaction together, with the same amount, and therefore itâs hard (at some point impossible) to say, what the history of your UTXO is.
As long as you stay off-chain Cashu seems good, but on-chain it doesnât really change anything imo
Well it would if I sent you money on cashu and you redeemed it. It makes it to where the bitcoin never moved anywhere on chain and the bitcoin has completely changed hands undetected. It makes on chain surveillance almost useless because any bitcoin on chain could have changed hands and no one could know.
