I initially read this book not too long after it had come to public attention that the Everquest economy was the 17th largest in the world (or whatever), when virtual worlds were gaining public attention in a new way. What I had remembered about the book from my first reading was that it was an easy and mildly insightful steady walk of a read that would offer a nice break and change-of-pace, with a strong voice. I was not disappointed: it has a voice that's engaging and full of personality, sometimes but not always likeable, occasionally funny and sometimes only making the attempt. It is always easy and accessible, but can go on too long.
The voice that "Designing Virtual Worlds" has is that of the well-worn insider, the pioneer who has seen his mistakes recapitulated but also his triumphs amplified, and can no longer enjoy participating as its workings are too easily perceived. Richard Bartle is a man defending the choice of his career, but mostly to himself, doing his best to leave our judgment open, instead informing us with his perspective. He broadly scours the academic literature for anything that could be relevant, but has come away with populist tastes, favoring the social scientists who have spent considerable time in virtual worlds, whose scholarship has given them a slightly different perspective but no overriding theoretical underpinnings. In a very large academic survey, this book is oddly reticent to cover the author's home disciplines of CS and AI, or really anything but social science. As best I could tell, the view behind this book has come to the perspective that the socioeconomic realm is all there really is to the human experience. There is a huge, and very ironic, gap.
This book is interested in telling you what virtual worlds are about and deeply disinterested in showing you what they are. One could read this book and come away with no real idea of what a design of a virtual world actually consists of or looks like (that is certainly my experience). Perhaps for the better, it assumes that if you're really interested you're going to be playing a number of them and analyzing codebases and whatnot. Instead, this book tells stories about things going wrong and how interactions could work. It 'tells' instead of 'shows', trying to avoid prescription at the cost of occasionally obscuring what choices might even be. It proposes models of the different motivations of people playing virtual worlds, how those motivations interact, and how those interactions develop over time.
Here's what I took away this book thinking virtual worlds are about: virtual worlds are a means by which people can symbolically undergo the crucible of personal risk, sacrifice, and effort in the name of shared community, when their physically-bound social experience does not allow them any clear path towards anything like heroism or a community-rich interaction. The challenge of the design of virtual worlds is the designing of a setting of interactions where players can experience this heroic development without it being spoiled by others at different stages in their own journeys. A good virtual world will generally prepare its players to better cope with their socioeconomic reality.
In a way I did not perceive in my first reading, or even during the second, is that it's possible to read this book in a way it was never intended: embedded in a broader physical reality. This book's fundamental gap is having any awareness of our physical specificity: it presents ecology as being pertinent only to the degree that it affects the economy, physics only the degree that it matches people's expectations, etc. But what would it look like if this knowledge was grounded in our actual physical situation?
There is a radically different response to the problem that a player's true physically-bound socioeconomic situation does not allow them the opportunity to experience real growth under shared adversity, which is to do better at changing that situation by designing better real-world experiences. What can be learned about the personal development of players that could lead to designing real-life experiences that allow their heroic development to make contributions to real ecosystems instead of exacerbating the footprint of computational infrastructure? Once seen, the massive absence of grounded experience hangs over this book: we see the types of motivations of players, but not those of non-players.
The stakes are fascinating: what are we losing by people undertaking their character-building sacrifices not towards real ends? It might not be all bad: it would be nice if people could work through being griefers in situations as fake and isolated as possible. Given this, one wonders what a hybrid of real and virtual experience design might look like.
Here is the real intellectual gap of this book: design. This is a scholarly design book without a scholarly design background. It contains a design perspective that doesn't have a designerly way of knowing. It's a book about designing that doesn't confront designing. If it did, it would find the real question is not how to design virtual worlds, but how to design growth experiences.