Ultimately lightning will probably become the glue between different ways of “sharing” UTXOs, if all these different options can speak lightning they become interoperable allowing for instance payments with minimal on-chain fees. Just like the base layer isn’t for coffee payments, eventually neither will lightning. The fees associated with opening and closing channels becomes economically when the amount of transaction across those channels goes up a ton. If these channels are connecting lots of “shared” UTXOs solutions, they make a lot of sense. Right now it’s hard to see this as we’re still in the process of building out and deciding the best way to do “shared” UTXOs, cheapest way is just have an sql db and be a custodian (but is of course not ideal for all the reasons). fedimint is a step in the right direction which doesn’t require a base layer chain and gives added privacy and reduces rug pull risk vs having a single custodian, but there are still issues with that solution. There are other options in the pipeline that might be possible on the base layer, however, any changes to the base layer take time (which is a good thing). At the end of the day, the question becomes “are the fees I’m paying for opening + closing a channel worth the economic benefits I get from having it” which is a pretty reasonable incentive. Early days, lightning is still a child, at ~5 years old, compared to many of the tools we use in tech today.