But this data is actually useless without context. I actually have no clue if 500 calls since August is a lot -- to be honest my first reaction was that it sounded low compared to my (completely possibly incorrect) notion of how many uber rides must be taking place in New York (Uber's biggest market?). It would have been nice for him to have done some analysis with this supposedly indicative number, for example saying how that represents x% of all rides or something. Instead he focuses on really strange parts of this story. I suppose its an op ed so excusable, but I think there was an insightful possibility here that instead just comes off as some guy who's kind of angry and snarky.

Especially since it sounds like most of the calls were for people who didn't understand how uber works and wanted to pre-book a ride. I'm not sure that it would ever be worth having a listed phone number just so someone could answer and say "actually you can't pre-book, you'll have to buy a smartphone and download the app".

You could set up an automated phone tree for this. I've seen multiple large organizations have phone trees where several paths down the tree leads to an automated answer at the leaf (sometimes you can't even get to a representative once you reach some leaves).

Sure. You could. But there's a big step between it being possible, and being (what the author implies) a moral obligation. In fact, there's probably a big step between it being possible and it being worth it.

If Uber actually got big enough that it became one of the only transportation options, then they might have some kind of moral obligation to setup a phone tree to help people who don't have any other options. But I don't think they're anywhere close to that.