I'm currently sitting on the fence about how I feel about the benefits. As smooth mentioned, it can work out when there is bifurcation between BTC and HIVE, and given BTC's lower beta, would likely be less impacted by market wide drawdowns. Essentially, the delta between the two would be the surplus we would be able to capture and sometimes this can be quite large. On the other hand, given the proposed 1 million worth (a rather small amount compared to the market cap of HIVE), a market wide drawdown would still mean an overall larger drop in the market cap of HIVE backing the HBD, and only a tiny extra bit of buoyancy from the BTC reserve to help out.
I would also like to ask, if there are any discussions around adjusting the 5% fee for conversions from HIVE to HBD. I don't know the specifics and couldn't find the exact information, but I am under the impression that LUNA to UST does not incur the same kind of fee, nor the opportunity cost of 3 days time. I feel that this could be a deal breaker for many who want to deploy larger sums. I know that I would personally feel more comfortable and willing to buy more HIVE to convert to HBD if the conversion fee was more plateable but it remains an obstacle I'm currently unwilling to cross. I am aware that the 3 days is to mitigate potential manipulation and price volatility, but I am also quite certain that arbitrage via conversions in highly volatile environments is something larger players would be interested in, myself included.
All valid points.
I think the 5% fee is a bit to much, also there is no logic to have a fee only in one direction, why not in both, and maybe just a 1% (0.99 to 1.01).
About the 3.5 days, the one thing I find this useful is the price feed.
If you take a look at the t20 witnesses you will see that some updated their price, once per hour, others once per 5 hours, sometimes some do it once per day.
The 3.5 days basically allows better price oracle and security, and this is not to be undermined. I would like to see faster conversions but also better price oracles.
I'm not sold on the T20 witnesses operating this way being an effective way to deliver a responsive, and yet decentralised price feed. Understandably, the purpose of Luna is entirely stablecoins (which is ironically being undermined more and more by it's BTC holdings) and so our considerations are a little more complicated than just being ultra responsive. Perhaps if we had a greater number of participants, and several orders magnitude more participants / liquidity, then the 3.5 days can be reduced a bit.
In the same way that missing blocks has financial consequences, price feed updates probably should too. It is obviously a critical component of everything that is being discussed here, and currently gives no leeway to shorter conversion times for the simple argument that some witnesses update too infrequently. I'm not talking by the minute, or by the block updates.. something like every 8 hours minimum so that we can have 3 periods within a day to form the 3 moving average.
I feel like the price feed has not been discussed enough around here neither the rules around it. As it is now it is completely arbitrary. Any witness can set whatever price he wants at whatever timeframe. I feel like at least there should be some timeframe defined for witnesses to follow.