It's already got enough liquidity to handle quite a substantial rise in usage. The ongoing work by Lightning Labs with Pool, and Breez's LSP protocols is moving towards a situation where the number of channels and their added transaction capacity will drive the ratio down so that it doesn't matter so much.
Sure, DIY node running is gonna be more complicated but at the same time, if you trust your counterparties well enough and they have been reliable you don't need to put any tx's on chain. Improvements in the protocol will likely make this even better over time.
I was just reading about "sidecar channels" before, which are built on top of other channels. I'm sure by the time we see $20 bitcoin transaction fees again this will not be a problem for LN's capacity to cope with an influx of new users.
Yes, it has a lot of liquidity, but new people joining won't be able to use any of it, but will need to open their on channels, unless they go custodial, that is.
Sidecar channels sound good, yes, let see when they're ready...
reply
In many cases protocols that use LN micropayments under the board, there can be multiparty escrow schemes that enable users to pay via external channels (including custodial) that allow the purchase of vouchers for service, without the friction of waiting for their own channel to open to get started. It has other sticky issues, like how to automate that process, but I know for sure at least some LN wallets can be controlled using an API, closing the loop.
reply