Many zero-conf proponents work on the bleeding edge of supporting Lightning, including myself. Lightning is not risk-free and the base layer should not be assuming it as a primary dependency for commercial payments.

for low-value payments, lightning is the only workable version because the current low-fee environment is not scalable and never will be

for high valued payments, zero conf was never valuable or useful and never can be - it was always the beneficence of users you are relying on low fee/high fee double spends of a zero conf with no rbf flag has been demonstrated, repeatedly ad nauseum.

... so there is no long term scenario where zero conf is valuable.

right *now* with low fees it might "seem nice", but really it just incentivises network-wide surveillance, increased peer burden on nodes, and unsustainable practices that are akin to a "mev" bounty hanging over merchant's heads.

also, i've been using bitcoin for years without zero conf. selling and buying online. operating merchants with millions in transactions. the 20 minute wait before i ship is meaningless, and the only risk i take on is that a user replaces a transaction with a different destination. which i've never observed. even with the flag on (which i dont care about, and never have).

and if i do observe it ... i just won't ship. it was easy to code up. the user will have to initiate a new tx. i have no objection to a user being able to cancel their order. why would i?