[bisq-network/bisq] Haveno - Bisq fork from the Monero community (#5429)

[chimp1984]

I looked a bit into the [haveno project](https://github.com/haveno-dex/haveno) and wanted to share my impression/opinion.

They have forked the Bisq code a while back to evaluate a proof of concept to use XMR as base currency and it seems they got the basic funtionality working using Woodser's java rpc library for the Monero functionality.
They want to use a 2of3 multisig with an arbitrator who will be one of the team members. Monero has very limited smart contract capabilities. There is no timelock and you cannot created chained transactions. Funds from a transation can only be spent after 10 blocks/20 min. All that restricts the potential options a lot. The current protocol Bisq uses would requires a time locked tx so that cannot be implemented (beside that they don't want to use the DAO anyway). The maker fee tx and taker fee tx play several roles which become problematic in their context.
They currently plan to keep those transactions but that comes with usability costs (time) and higher failure rate of trades due the time delay.
A maker need to wait 20 min after creating an offer until it can be taken. That might be not a major issue if the publishing is just delayed or its publsihed immediately but cannot be taken before 10 confirmations. 
But when a taker takes the offer the taker fee tx also requires 10 blocks. Here its really problematic as the maker might go offline in those 20 min and then the taker has lost his trade fee and all ends up in a failed trade.
After that the deposit tx requires another 20 min (compared to 10 min in Bisq), so thats not a big problem but also not an improvement. And then after the trade is completed the payout tx will require again the 20 min before the user could use those funds for another trade. Doing multiple offers and trades in a row will be a time consuming task. 

I think they have to remove the taker fee tx at least, which does not cause too much troubles. Removing the maker fee tx would be more difficult as that serves to keep the utxo for the trade resevered and not having that requires more complex wallet management. Also not having a maker fee tx adds vulnerability for offer spam. 

My biggest concern though is their decision to go back to the trusted 3rd party model of a 2of3 multisig. That comes with legal and security risks and privacy and scalability issues.  
But I also do not see any other solution as the current Bisq model using the DAO instead cannot work with Monero even if they would keep the DAO.

I was considering myself to use XMR as base layer with a 2of2 Multisig MAD (mutual assured destruction) model, but came to the conclusion that this is too risky and would not find acceptance. What I have learned recently about the limitations of Monero reduced my hope that there is a viable solution using Monero as base layer. A pity as the strong privacy and low miner fees would have made it a perfect candidate.


-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/bisq-network/bisq/discussions/5429
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bisq.network/pipermail/bisq-github/attachments/20210421/4af609d5/attachment.htm>