It's not about trust, but about the potential mess caused by lost transactions.

In the good old days one of us could buy a round of beers in the bar, and everyone in the crowd could pay for the beer through their phones - it was easy and it was fire and forget.

Not anymore, nowadays the sender and recipient have to baby-sit the low-fee-transactions, and one almost needs to be a bitcoin expert to be able to send and receive payments with a proper but not too big fee paid, choosing weather to use or not to use RBF, eventually try to get the transaction "unstuck" through CPFP and other means, rebroadcast a timed-out transaction, etc.

Yes, lightning will eventually be good - but I am quite concerned that it's overhyped, hailed as a silver bullet while it's not.

For one thing I expect it will probably be at least a year until a majority of wallets supports lightning.

Further, the two parties needs to be connected via a payment route. I can easily see that it will be easy since "everyone" would like to connect with the biggest hubs, but then again, there is the centralization risk.

Also, it's needed to make one or two on-chain transactions to move funds into the lightning network, this may eventually become very expensive if we'll stick to the 1 MB block limit.

Finally, one has to decide weather to store the bitcoins in a payment channel or on a private key; every movement in or out of the lightning network (as far as I understand it) involves on-chain transactions (and it may be easier to move funds into the lightning network than out from it, I believe?).

I think Lightning is not so different from any alt-coin. It's fully possible to move value measured in bitcoins cheaply through ethereum, bitshares, ripple and probably several other alts.

Please correct me if I'm wrong, I haven't read the whitepaper.