Disagree. I have believed in the potential of networked payment channels since Mike Hearn's work circa 2013. Properly applied, it makes perfect sense, economically and technically.
Networked payment channels could be a great supplement to Bitcoin. They're useful for very small microtxs (on the order of <1cent), especially recurring txs, e.g. paying for pageviews.
Imagine the LN, if nobody ever dreamed of using it for moving more than a couple of cents. There would be no liquidity issues, ever. Anyone with $10 could be a "liquidity provider"...
High-degree, high-liquidity hubs would be far more decentralized. You wouldn't need watchtowers, since the risk of channel theft would be meaningless when channels contain just cents.
Now couple that with low on-chain fees, so channels could be opened and closed much more easily and frequently, and the risk/liquidity barriers become quite negligible.
Incentive for what? For providing liquidity? Liquidity in payment channel networks is a non-issue. It only seems like a problem because the BTC people tried to use LN for txs of >$10.
Incentive for preventing channel theft? Almost irrelevant, assuming that being an LN hub is a repeat game, and every channel never holds more than a few cents.