The calculations in the article seem to be based on just a couple numbers -- such as the 40,000 miles per year that Uber gave for the purpose of cost of ownership calculation, which is used to try to calculate the hours drivers work and the fare they earn per hour. Some real numbers about fares earned would help here. The conclusion is not clearly stated in the "Conclusion" section, but here's my interpretation: If you make certain assumptions about an uberX driver, you find they make about as much as a typical taxi driver. This suggests that driving uberX instead of a taxi does not automatically put you in a completely different income bracket, but I think the devil is in the details. Obviously the spread between different drivers is much, much larger than the spread between taxis and uberXs. If we made a histogram of income for each and superimposed them, they would be two humps with a lot of overlap. It would be more interesting to know, for example, how the jobs of top-earning uberX drivers and top taxi drivers compare.
Then there are implications that Uber is another Groupon, and its financial success is dependent on a steady stream of investment dollars. Well, Uber is making plenty of money. They write software and do operations and the money just pours in. Every user is a paying customer, and the vast majority of users come away satisfied and happy to have a better way to get from point A to point B. The drivers make money and are thrilled. Some people take Uber literally every day. Seems like a pretty good business to me.