An information good is one for which it's very cheap to produce additional units. Unfortunately, I'm probably abusing the term, since books are listed as an example of information goods. The idea is that most of the cost of producing the book goes into the content (writing, editing, etc.) and actually printing one additional book is cheap. (When I said books are "imperfect" information goods I meant that they still require paper and shipping and retailing and such, but it's probably just a bad use of the term.)
There's a fantastic book about the economics of information goods called "Information Rules" ( https://www.amazon.com/dp/087584863X ). Software is an information good so this covers some topics relevant to the digital economy. My favorite part is the chapter on lock-in. In particular the discussion around the equation:
profits from a customer = quality advantage + switching costs
which puts "(marginal) goodness of your product" on equal footing with "pain you can inflict on your customer for leaving".
https://www.amazon.com/Information-Rules-Strategic-Network-E...
I know it may sound not helpful but achieving vendor independence is a valid reason for adopting at least some software design patterns so you can swap the implementations without changing the interface. It’s becoming increasingly more important as the cloud service providers are getting more aggressive just because they can get away with it.