Pete On Software

RSS Feed

Archive for April, 2008

Y Kant Developers Read?

Bookshelf image from
I guess there are two major reasons that developers don’t read many books. This is kind of a timely subject for me right now on both fronts.

First, I began a new job at the beginning of last month. My new employer was running a Visual Fox Pro environment and wants to move to a full-blown .Net SOA Architecture. I was brought in to architect that. The team that I have (save one programmer that I recommended and was hired) are all Visual Fox Pro 6 developers who don’t know a stitch of .Net. With the exception of one guy (who I *really* appreciate), none of them seem to be very interested in reading books to find the answers to problems or to learn the gist of C# and ASP.Net.

This was also the case at my last job. With the exception of the guy who I recommended at my new place, no one would read books. They would take them at our insistence and then never read them. Someone once told me that developers read about one technical book a year on average. I couldn’t believe it when I heard it, but I’m starting to wonder if that number isn’t smaller. The first major reason just seems to be laziness and/or apathy.

The second major reason seems to be the driving force behind Jeff Atwood and Joel Spolsky‘s new venture, StackOverflow. According to Joel, “Programmers seem to have stopped reading books. The market for books on programming topics is minuscule compared to the number of working programmers. Instead, they happily program away, using trial-and-error. When they can’t figure something out, they type a question into Google.”

I believe that to certainly be true, and for people who are using new technologies or are on the bleeding edge, trial-and-error is the only way to go. However, if we are talking about anything that is approaching a year old, books certainly exist. My problem with trial-and-error and Google is that you never really *learn* anything. You are kind of limited to your own cleverness.

Advanced books not only teach you new features, they teach you new techniques. If you just trial-and-error, you will only try to figure out the Python way to do your C# code, or the Ruby way to do what you’ve always done in Java. You won’t learn what there is to learn about the new languages and the cultural mindset that comes with them. You could skip reading books and read lots and lots of source code, but I don’t believe a majority of developers are doing that, either.

However, so far we’ve only talked about learning a new programming language or technology. There is no way to effectively trial-and-error the kind of knowledge that you get from books like Code Complete, Pragmatic Programmer, GOF Design Patterns, or Practical Cryptography. How in the world has this style of learning, proven effective for centuries, fallen so far by the wayside? I am not trying to seem terribly elite, but I really read more than an entire technical book per month on average, along with several technical magazines and journals. I directly attribute my success in my field to that fact. Why can’t people see books as an avenue to their greater success?