[smcmurtry]

I've often thought it would be cool to build a fairer ride hailing app that gives drivers more autonomy. The driver buys the app as a one time purchase, they get to set their own prices, and there is more transparency between buyer and seller. There could be a simple bidding process where users request a ride, drivers make an offer, and the user accepts one based on price, how far away the driver is, and their reviews. Basically the app would be more light-weight and be more like a marketplace.

[saas_sam]

Sounds great until you realize 1. riders would only use the app if they could sort by 'price,' 2. drivers would therefore have to constantly change their rates to reflect what those in the area are charging at a given time/supply/demand level, so.. 3. in order to do this effectively without creating massive unexpected price swings for both drivers & riders, you'd end up automating this 'bidding' system and hey whaddya know, you just built Uber again!

[smcmurtry]

Genuinely curious about this - why would the price swings in this case be worse than with uber? With enough drivers, I would expect prices to reach an equilibrium that depends on time of day/day of week, with highly rated drivers charging more. And even if the price swings were larger than uber's, wouldn't the prices be more optimal since they would be set by individual actors with more local info about the cost of providing the service?

[xocyabencl]

While Uber and Lyft raise prices in periods of high demand, they also subsidize rides in periods of low demand to keep a consistent quality of service. You'd have a hard time getting a network as reliable as Uber and Lyft without that subsidy.

[AntAF]

It's not uncommon to see Uber retain 75% of the fee collected these days. The subsidies are much lower and less frequent than you are probably expecting. In Los Angeles, you are talking about 1.1-2.0 subsidy on base rates in the same areas with base rates for the driver being .72/mi and .11/min.

[nugi]

So, decentralized and trust based. But then the challenge is to engineer around abuse on either side. Seems a worthy project.

[beojan]

That sounds like a great idea, the only point I'd make is that a one time purchase model wouldn't work because you'd have ongoing costs (transaction fees for payments, records you legally have to keep, etc.).

[smcmurtry]

Good point - a subscription model would probably be better.

[mcrad]

I want to make this and include a multi-level marketing network model so drivers are additionally incentivized for recruiting and retaining halfway decent drivers. Herbalife for transport!

[killerstorm]

In Ukraine we have a taxi app where user selects price he's willing to pay and waits for drivers. He can increase the price if no drivers take an offer.

[barbs]

Reminds me of a similar project that popped up on HN a while ago.

https://arcade.city/

[DCoder]

Semi-related: there's a startup in Denmark [0] trying to do this for parcel/freight transportation. I'm not sure how well they're doing, though.

[0] https://www.badabring.com/ (link in Danish)

[SilasX]

That’s basically what SideCar was, as drivers could set their own rates, and passengers could choose from a number of offers based on total price, quality, driver rating, and time until pickup.

SideCar closed in late 2014 though :-/ (And there was a recurring cut paid to SC.)

[boyce]

Something like what Project Wonderful does with web advertising would really have big potential