I sold 467 copies over 10 years for less than $5 per copy. It's about $2000 in revenue. Surely not a good use of my time. But I liked the app and so did its few users. We should build things we want to use ourselves right? I think it was a much nicer and faster frontend to the Yelp restaurant directory than the Yelp website.

Your other question about the cost is answered in the blog post and in several other comments I have left in this thread. In short, when I started the app 10 years ago there was no paid API (perhaps there was some kind of "enterprise" version but I'm not sure). The point of my post is how poorly they handled the transition from free->paid.