Problem Statement
Although Uber initially offered drivers and riders an interesting new alternative to taxis, the company achieved monopoly status and has at this point nearly killed all rival services, allowing it to charge whatever it wants. Whenever I take an Uber I ask the driver what % they are getting of what I am paying, and (in San Francisco at least) the drivers are nowadays getting around 40% of the price I am paying.
This seems highly inefficient, to say the least. There are many millions of rideshare/taxi drivers worldwide. Could an app easily be built that would connect riders with drivers, and allow the drivers to re-capture the majority of the fare?
Requirements
A ride-sharing app needs to have three key components:
A payment mechanism, allowing a rider to quickly pay a driver
A reputation system, letting drivers and riders rate and trust each other
A signaling mechanism, for riders and drivers to find each other
Payment Mechanism
This is pretty easy nowadays. For a MVP, without writing any code at all, the drivers could simply setup their own way of collecting payment, such as Apple Pay or Stripe or whatever makes sense for their city. The fees paid would be a couple percent, as with buying a coffee or sandwich.
Reputation System
This is the most important and difficult piece, because drivers and riders need to trust each other and stay safe. The way to do this is to allow drivers and passengers to rate each other without any social pressure that would bias reporting, and then let them lookup a rating weighted by people they trust. The way it would work:
Both rider and driver mutually share a public key with each other before either accepts the ride.
Each then searches only for reviews created by other individuals whose reviews they personally trust. This is the key innovation. There is no global source of reviews, because that could easily be gamed by bots creating fake reviews. Instead, riders and drivers pick people they trust and consider only their reviews as valid. Because people are connected to so many other people (the classic ‘six degrees of separation’ stuff), trusting the reviews of a few hundred other people will give you information on others.
Signaling Mechanism
The critical aspect of the signaling mechanism is to preserve privacy so that you cannot cause harm by accessing another person’s location. This can be done by a two-stage process:
First, the rider’s app posts to a public server a general area in which the rider is located, but not specific enough to find them…. say a few city blocks radius in a city, or a few miles in the countryside. The posting simply says “someone within this general area needs a ride”.
Then, an available driver responds to the rider with their public key (and a signed proof of that key). The rider looks up the reputation of the driver (filtered as described in the prior section). If the rider trusts the driver, they send their own public key and precise location to the driver, who does the same reputation check and then accepts the ride.
People are already working on this
Additional proof that this can be done is given by the fact that other companies are already working to launch this sort of service. One that I’m aware of is Teleport, who I think is already offering service in a test city. But the opportunity for creating public good is so considerable that I would encourage others to think about working on this. And if you are interesting in working on this at my lab in San Francisco, please consider reaching out to me.
As a person who just spent two weeks daily using Uber heavily for a consulting gig, I found no trust in the Uber system itself; and was dismayed at the small amount the drivers were actually getting.
While I do not have the acumen to set up such a system, I do believe there needs to be others the the 2 "big names" who are currently monopolizing the field.
Thank you for the heads up Philip.
Interesting information, but it seems to me that it's also about the customers themselves. You can create an application that will be many times better, but people will be skeptical about it, because of distrust or misunderstanding. Most users often perceive something new with hostility. It is easier for them to use what is convenient, comfortable and what they are confident in. And the monopolist company takes advantage of this by changing the conditions that are suitable for it. Since there are no analogues.
Therefore, I believe that a new product requires either an innovative approach, something that other competitors do not have, or the creativity of the product. And also a big marketing campaign, since no one will want to use an unknown thing, no matter how convenient it is. Unfortunately, the realities are as follows.
I tried to tell you about my opinion, as a direct customer of the product :)
Thanks a lot for the information Philip. (Could you get in touch with me? I would like to discuss some things with you.)