This seems to echo and reinforce a point in the article. The apps aren’t replacing the delivery person they’re replacing the logistics management with you the consumer and letting you be mad at yourself, the driver, the restaurant, and the app in some set of proportions when net the balance of ire should fall on the app for the state of affairs.

You realize that absent this, you still have the issue of not getting your food? If there are more people wanting deliveries than there are drivers, you'll still be left waiting. This merely gives you more options (eg. "delivery within 10 minutes for $15, delivery within 60 minutes for $5), compared to the alternative of the platform trying to make a one-size-fits-all price for everyone (eg. some sort of surge pricing system).

I interpreted them to mean that if they are going to use the app instead of cooking or eating out, they just want to pay to get food. If there is uncertainty about whether the food is coming or not, there is no reason to use the service.

> If there is uncertainty about whether the food is coming or not, there is no reason to use the service.

And what I'm saying is that there's already uncertainty. The current interface of DSPs showing that it'll arrive within 30 minutes or whatever, is a leaky abstraction of how the underlying system (ie. restaurants and delivery drivers) work. Just as how DSPs likely have some sort of system to estimate delivery times based on current supply and demand, they can build a system that also factors in price, and give consumers various price options and their associated estimates.