In this post and most of the comments here, people assume you have a binary choice: 1. You can put your book online for free and sacrifice extrinsic reward (mo

In this post and most of the comments here, people assume you have a binary choice: 1. You can put your book online for free and sacrifice extrinsic reward (money) for greater intrinsic reward (popularity, impact on the world). 2. You can maximize your financial income by only selling the book. I thought I was making that choice when I decided to put "Game Programming Patterns" online for free, but that's not actually how it turned out. I have made much more money from the book than I would have if I'd gone with a traditional publisher and not put the book online for free. The model I have now is something like the classic marketing funnel. To make the most money from a thing, you need to have as many people go through this series of steps as possible: 1. Know that the product exists. 2. Decide to want it. 3. Be just willing to pay the price to acquire it. (Anything less and you're leaving money on the table.) 4. Keep as much of that money for yourself as possible. (In other words, reduce costs.) For technical books, most people don't get past step 1. Putting the book online dramatically improves that. Of the people who do, many stop at step 2. There are so many articles out there, it's easy to convince yourself you don't need a acquire a whole book. Again, putting the book online helps: you can try before you buy. Step 3 is the interesting one. Putting the book online for free obviously leaves, like, all money on the table. But what I have found is that there a self-selected market segmentation seems to come into play. Many people do read the book for free, but some choose to pay anyway. Step 4 is the dirty secret of the big technical book publishers. They take an embarrassingly large chunk of each sale. That could maybe be justified in the old days of publishing when you needed to run a whole printing press and maintain relationships with independent book publishers and all that. But, frankly, they do not do enough to justify how much they take. Self-publishing fixes that, at the expense of having to do more of the work and management yourself. I had to find a freelance copy editor, and typeset and design the cover. But I effectively "got paid" to do that work at a much better rate than it would have cost me to "pay" a publisher to do that by giving them the lion's share of each sale. I don't want to generalize too much from my one single data point, but it seemed that for me, the increased widening of the first two steps and the greater share at step 4 more than compensated for the money left on the table for some readers at step 3. It was a clear enough signal that I'm taking the same approach with my second book. And, equally importantly, I really liked the subjective experience of it being my book and being able to put it out into the world exactly the way I wanted to.