This discussion is very helpful. Thank you Can't the eventual fractional reserve worries at least be mitigated with the "automated banks runs" @calle has proposed? My left side of the curve understanding was the ecash has an expiration of say 1 year, so you have "epochs" of ecash that all expire at a certain point and need to be rolled over periodically. Thus you have a regular interval of "bank runs" that help keep the Fedimint or Cashu mint honest. While certainly not perfect this seems like a strong mitigation. I guess in theory they could still do shenanigans like frac reserve into short term interest bearing products that expire in between the ecash "epochs", but something like that at least becomes much harder, more likely to be caught, and more costly. With a Fedimint or Cashu mint maybe that 1 year time period could also be adjusted for different groups with different threat models and tradeoffs. Maybe a hardcore security minded Fedimint has a very short window for epochs say 3 months, and the tradeoff is more overhead and work for the guardians and people have to come online more often. I would way rather pay a team of guardians 20bps in fees on my Fedimint ecash balance that has a robust security model with short term regular bank runs than an ETF. Maybe its like an Uncle Jim ETF option? Or maybe even the epoch/expiry time periods could be somewhat randomized to make it even harder to frac reserve, e.g. this epoch has a 9 month expiration but the next one has a 7 month expiration? I'm not a dev so maybe that's way too complex. Just trying to brainstorm. I think this is either a solvable problem or at least one that is mitigate-able. If ecash had automated bank runs and I knew and trusted the guardians the tradeoffs and security model are plenty strong for my tastes long term in small to medium amounts and not my core stack.

I'm trying to fix that: https://gist.github.com/callebtc/ed5228d1d8cbaade0104db5d1cf63939