Dreaming in Code, by Scott Rosenberg

I actually read this book nearly a year ago, when it came out in January, but wasn’t sure how to do a review of it, since I’m so involved with the subject matter (I’m on the Board of Directors at OSAF, and helped Mitch start it several years ago, while I was at Reactivity and he was an investor there). I have too many personal reactions and feelings about the book to write a very useful review, I think. My short analysis, though, is that it’s definitely not a “modern version of Tracy Kidder’s Soul of a New Machine,” as some of the reviews claim. Kidder’s book about Data General and the techie culture of Route 128 was a seminal achievement, and one of the first really useful and understandable accounts about how techies do their work.

This book tried to give insight like that into the software industry, and at the same time tried to describe open source dynamics, and I think it was only marginally successful at each. It also illustrates the authorial hazard of trying to write history while it’s happening — when he was choosing his subject matter, it seemed likely (to some/many) that OSAF would be the super-successful open source project for consumers — Mozilla was a bit of an afterthought at that point — but we know how the Mozilla story played out, while OSAF is still a bit in the middle.

I think many of his conclusions (software is hard, for example, or that complexity is outrunning our ability to cope with it) are too heavily rooted in the particulars of OSAF, and he would have had a completely different outlook, and made different conclusions, if he had been at Mozilla or Google or Facebook over the past few years.

It’s an okay book, but I think has the serious flaw that he was trying to write about history while it was being created, and lacks a sense of perspective as a result.