It is reckless behavior to not have synchronous or asynchronous replication for data that cannot be lost.
Most LN node operators do this but somehow not a large custodian.
Most node runners don’t support a UI, users, nearly the same amount of bitcoin, a separate database for the custodial balances, all sorts of features, etc. It’s easy to look at one thing that goes wrong and call them reckless, when in reality they are doing an amazing job and have solved 500 other things that could go wrong
2🤙1
