Even if agree with you, centralized vs. decentralized makes a lot of difference, this is not the main point of the article. Centralized vs. decentralized influe

Even if agree with you, centralized vs. decentralized makes a lot of difference, this is not the main point of the article. Centralized vs. decentralized influence your go-to-market strategy and probably your probability to succeed, but not directly your business model. The point made in the article is that the biz model of the "Uber for X" businesses is not very strong unless X is a heavily regulated market and you are able to skip part of the regulatory burden to provide a far better service. What's missing in the article, is the effect on the market size. Uber and Lyft changed the market size. I've taken probably a cab five times in my life before Uber existed, probably I'll use Uber five times before this week ends. And the same goes for many people. What about SpoonRocket, for example? Well, I ordered food delivery before it existed. Probably having an app and many more restaurants to order from increased my consumption of food delivery service, but just a 10 - 15% more. Not enough to sustain new players and make enough people switch fast enough to build a sustainable barrier to entry. Not sure I'm clear: Uber provided a service so much better that increased greatly increased the market size and captured a lot of new users (and old users were fast to switch). Other services failed to increase the market size and didn't provide existing users enough benefit to switching, so no barrier to entry for competitors, a lot of competition, a lot of startups that fails. I have a doubt, though. We are seeing a lot of Uber for X failing, but that's because a lot of them were competing for the same (in the US car-hailing market there was just Uber, Lyft, and Sidecar). That doesn't mean that one or two company will survive in the end.