Hey Will, thanks for this, and I hope you have a good start to 2018, likely to be an exciting year for both 0x and AirSwap.
>The overhead associated with negotiating each over-the-counter (OTC) trade makes this approach unsuitable for moving small amounts of value.
Peer-to-peer trading is very popular for trading in electronic markets, for example corporate bonds, penny stocks, CFDs, and FX.
> [AirSwap] Liquidity providers must be online and available to negotiate each trade. Cannot fire and forget orders.
You are right that limit orders will not be supported on the first iteration of AirSwap. In my years of trading at Virtu, we were market making using bots on a 24/7 schedule. This is true for both order book and peer-to-peer markets.
>[AirSwap] cannot source liquidity on demand. There is no guarantee that a counter party is available and interested in trading the same size or at the same price point and the only way to find out is to engage.
This is true of order book models as well.
>[AirSwap has] no networked liquidity. Results in a closed liquidity pool that isn’t directly accessible to other exchanges or dApps.
Any market maker connected to AirSwap will have the ability to connect with any market taker on AirSwap. In the long run, these will likely be the same makers connected to 0x relayers.
Open relayers in theory can share orders, however I suggest considering what collisions and front running concerns this further introduces.
I’m glad to see you acknowledge the front running issues and are actively tackling them. I look forward to seeing potential solutions in parts 2–4.