The easiest solution I can see is essentially setting alert thresholds "tell me when it reaches $30/hr and I will consider it on the weekend".
I never used either side so I don't know their implementation.
It’s been pretty interesting.... the rates have gone up but not as high as we thought.
I know an independent limo driver who negotiates his own rates. It's not an efficient system and he has very little visibility into the supply/demand curve. And it's hard for him to drum up new business.
They would probably take _some_ pricing risk, since they might not want to "lock" drivers that they've used as the price basis, but that seems like a minor problem: Uber employs lots of engineers and I'm sure a few of them can figure out a way to safely manage this pricing risk.
I don't really see how this could work. the first issue is getting the pricing information from the drivers. are they willing to set a flat per mile rate, or are they going tailor fares on an individual basis, after browsing the currently requested rides? the latter would make it impossible to just compute a price from querying a db; there might need to be multiple rounds of bidding before the rider and driver could settle on a price. the former case would not be quite as complex, but uber would still have to ask the customer how they value wait time vs. price or decide for the customer.
i personally like the model of bidding on rides, but most people use uber/lyft because it's as simple as opening up the app, committing to a ride, and hopping in after a few minutes. i don't think it can work that seamlessly if every driver sets their own (possibly complex) fare scheme.