As most people know, this book by Scott Rosenberg does the "embedded journalist" routine with the first few years of Mitch Kapor's "Chandler" project. Scott of course expected Chandler to be a bit farther along after by the time he actually had to submit the book, but in many ways it just highlights the point about software development and its struggles. As has often been noted, this book joins the ranks of other outsider-looking-in books exemplified by Tracy Kidder's 'The Soul of a New Machine', which I thoroughly enjoyed when I read it aeons ago.
This book has several different audiences, each who will get something different from it. For advanced programmers, it's a light but easy read, providing one in a series of warnings about our general ineptitude at creating large scale software, something it's healthy for us to be all periodically reminded of.
For newer programmers, or those immersed more on the pure tech side, it's a healthy look at how people and processes make a difference. It's also a good introduction to some recent history which is often sorely lacking, and identifies and highlights some key figures in the discipline (e.g. Alan Kay). The book has a few digressions away from the Chandler team that really do provide some broader perspective (e.g. where did the software crisis come from, software engineering vs. real engineering, how do other companies llike Google or 37 Signals approach development practices, and more).
One notable digression compared how people are taught in a CS degree vs. a fine arts degree. In the latter, students study existing masters, learn about the history of the discipline, and work, critique and rework their own creations. Not a lot of that in CS, which generally teaches techniques in a context-free vacuum, with little perspective, historical or otherwise.
Given so many developers don't regularly keep up with trends in software process (through books, blogs like Joel on Software, etc.), it's unlikely the people who would benefit from this will actually read it, but what can you do.
Finally, for interested and slightly motivated people outside the industry who you've tried to explain what you do at work (parents, spouses, etc.), it may go along way to some better understanding of what you do. Or why you come home so frustrated some days.
In some ways, this was an incredibly bizarre book to be reading — as a book. You always think of books as chronicling some after the fact series of events, providing that enlightened retrospective. With this project both recent and ongoing, it just feels weird reading a book about it. Whereas if it were a series of blog posts, ... McLuhan strikes again. Part of this is that like many others, I'd followed Chandler early on. Some of it was the people involved, many who I knew of or had heard speak, and partly was the promise of both the applications and infrastructure they'd produce, along with the potential improvements in existing tools like Wx that they'd chosen to use. Again like many, my interest died out as the project floundered, but given it's at the point where they're actually starting to release stuff, the book may generate some new interest for the project.
So all in all, while the book will be nothing earth-shattering for most people, it's very much a well written and engaging addition that stands up well alongside others in this genre.
Comments