I asked Adam about the logic for it last year, he told me the fees go to pay for rotating the federation’s onchain utxos, to refresh the timelocks
if they dont get regularly rotated, eventually the onchain funds can be spent by a smaller emergency keys multisig
it’s a normal cost for any multisig thats using contracts with timelocked emergency spend path. basically every corporate structured multisig works this way, see anchorwatch contracts
honestly, big benefit of the way LN contracts are constructed is they dont need onchain refreshes to stay valid over time
iiuc blockstream has basically been eating this cost, maybe at some point tx fees onchain will cover it + help payback what they already spent
definitely a good question about incentives to run sidechains tho — unless you’re taking margin on swaps to/from the chain or issuing a token there isn’t really much incentive to run one, afaict
LND force closure SUCKS that bottom line
telling someone to RUN NODE selfcustody for hold 1000 SATS - losing 100000 SATS in channel CLOSING is BULLSHIT PROPAGANDA of certain maximalist here
that seems more LOGICAL answer to accept