Aside from e-journal downloads referenced in the article, I'd guesstimate that at least 50% of the coursebooks I used over the last year were electronic copies from services like IBUK (Polish online textbook repository), Springer or Wiley (and some other that escape me at the moment). It's often more convenient overall, and I don't need to worry about late return fees or lug them around, hoping I won't lose or accidentally damage them. Most importantly, when I need one, it's always there, and I don't need to sign a waitlist that's sometimes three years long for a rare book with maybe one or two relevant chapters. If the book was published after 1990, there's a stupidly high chance that I can download/access it legally from a computer. Also, you can do a word search in a PDF, something that physical books sorely lack. Programs like Okular even include tabs and bookmarks with comments/keywords, making the review process that much easier.
This also works for books that are so old, they're now in the public domain. There's so many interesting old books out there, not just fiction. Old scientific journals and textbooks, collections of sermons and speeches, engineering and repair manuals, on and on it goes. What's really interesting is, there's a lot out there that hasn't been digitized yet. I know two different people that work for two different private libraries that have literally hundreds of books each that date from around 1600 today and I bet the majority of the books in those collections can't be found on Google Books or Project Guttenberg (I know because I've poked through their indexes and wrote down titles to books that look interesting but I wasn't allowed to access because they're too rare, fragile, etc.). It'd be interesting to see the process involved in converting old texts to a digital format.If the book was published after 1990, there's a stupidly high chance that I can download/access it legally from a computer.