I recently read The Cathedral and the Bazaar by Eric S Raymond.
It's interesting to get a glimpse into the early history of arguably the most successful open-source software project there has ever been.
It's also dated somewhat. The author argues that open-source models of software development will overtake closed-source. It's clear today that open-source software has made enormous progress. Linux is more popular than ever. However there remains an enormous amount of closed-source software in the world 25 years later, much of it needlessly so. There is also just an enormous amount of software in general; I'm not sure if Eric in the late 1990s and early 2000s would have predicted just how central the software industry was to become in today's society and how much impact it would have.
I think the web took off in a way that the author may not have predicted. The other conspicuously absent topic is mobile. Combined I think these have significantly changed how most people consume software, and also how most people buy software. I think the author was correct that a lot of software would change to a subscription model; perhaps what he didn't anticipate is the prevelance of end-users directly making those subscriptions for software for their phones via app stores run by gatekeepers - the latest giant software corporations.