For me this has always been a counterintuitive conclusion as my own consumption interests are highly specialized and set, which is exactly the opposite of what Joanna (and by extrapolation, most consumers) are experiencing. I have seen some good peer-reviewed studies confirming this behavior and have lots of personal experience talking to friends and watching my wife consume media. In almost every instance consumer behavior is exactly the opposite of my own fixed and specialized interest. When Warner Brothers backed out of Netflix consumers simply shifted to other shows, movies, etc. This is exactly what Joanna means when she states she "cares less about consuming media that isn't there." What surprises me is that this is holding true even for books, which typically require a bit more time from the consumer to "taste". By this I mean that you can watch 5 minutes of a 30-90 minute show and judge your interest in it a lot more accurately than reading 5 minutes of a 300 page book. But Joanna indicates that this is exactly what is happening for her.
On this regard public and academic libraries have radically different patrons. Academic libraries have patrons with very specific needs and interests. On the surface public libraries may seem similar, but if what Joanna describes is true this means that breaking out of (or enhancing) the bestseller model that currently dominates public libraries may not be as difficult as I had imagined.
In many ways I think it failed at Douglas County Public Libraries where we put a lot of nonbestseller ebooks into the collection and saw very low circulation. As Kindle Unlimited is demonstrating, this is primarily a problem of marketing and delivery. Libraries have siloed, click-intensive interfaces that make it incredibly difficult for users to get to the nonbestseller in the first place. If we remove those barriers and bring the content to our users in a more intuitive, user friendly interface it is possible that we can get them listening, viewing and reading local and regional material. We just have to leave our current architecture and delivery model.
Update: Some hard data was just published that confirms this effect. Since the Big 5 Publishers won their battle with Amazon to set their own prices (I doubt if anyone is surprised that those prices are higher than they were when Amazon set prices), sales for Big 5 ebooks have fallen. And indie/self published sales are going up. It is a really long report but worth a full read. Here is one of the best graphs (the trend is glaring):