Wrt transaction times, if I would want to buy DOGE using BTC, and having BLK as collateral, wouldn't the slow BTC confirmations still dictate the swiftness of the trade no matter how fast BLK confirms? i.e. the smart contract would release the collateral only after BTC has gotten its first confirmation (or however many confirms the users have configured)?
In theory DRK isnt trustless because you can still track the blockchain mixer by watching the outputs. I think it would be fairly easy to write a script to reverse it.
If it was that easy, it would've already been done. Believe me, people are trying. It's a whole lot more complicated than a mixer which you can observe, but it's off-topic anyway and too large of a subject to cover here.
Hell, if they really wanted anon, they should try zero knowledge proof. Thats like the holy grail.
It's proven to be hard (impossible?) to set up trustlessly (to prove no more coins can be created out of thin air by those who create the setup parameters) and/or without the proofs getting huge and crippling the performance and scalability.
You point about Doge/BTC in a contract secure by BLK is 100% accurate. Doge is slowed down because it had to wait for the slow Bitcoin confirmations. Sadly, there is no way around this. However, i think we can use give the users an option to use a 3rd party service that checks Bitcoin mempools to have faster Bitcoin confirmations (this can be if both parties have agreed to it). That only works on Bitcoin because Bitcoin is very centralized.
Regardless, if you trade Node/Doge then the Node confirmations would be instant and they would have to wait for Doge. However, NightTrader does everything for you automatically! It loads all 3 wallets, it watches for all the transactions and it automatically fulfills the contract. So as long as you leave your client running, your trade should normally take no more than an hour(depending on how many transactions). IF for example you chose to trade 1 BTC for 1 BTC of Doge under a BlackHalo NT contract, we have to know how many transactions it needed to fund. Lets say it funds .1 btc at a time. So the BC contract would create a .2btc deposit contract and trade .1 10 times. So it would take about an hour.
However, if you traded BitBay for BC under the same contract it would take 10 minutes. So faster currencies would fund really quick in NT. And besides it is worth it, decentralized exchange with no trading fees and no middle coin!
You may be right about DRK coin mixer. Even Tor had security issues though so we know that security will always need improvement. As for Zero Knowledge proof, your point about "impossible to set up the network" is true for non-interactive proofs such as Zk-Snarks. For interactive proofs, then that trust issue isnt required. To make it "non-interactive" trust is required for the person who set up the network, they would know the secret and could take advantage of that knowledge. Additionally, interactive proofs would eat up a lot of bandwidth and processing time. Zk-Snark is much lighter in weight.
Actually, the bigger challenge with zero is simply figuring out a protocol!! Think of how many questions need to be asked to generate a proof. Aka "Is it a wallet on the network?", "Yes", "Are there coins in it?". "Yes"... and so on for each programming step you want to hide. Its intense. However, I do think it CAN be done and done efficiently. Unfortunately, i dont have the time to work on zero.
The best ANON is still in Halo/NT because when I send you DRK(especially with the mixer) and you send me BTC there is absolutely no way to prove that was done because its two totally unrelated networks. There is no link from A to B and because no exchange was used its not possible to know the users who performed that trade. At best, you would need to collect data from the Markets to somehow try and solve who was trading at that specific time (but actually, the same attack can be used on interactive zero).
