Citing: two extremes

In the bookshelf beside my desk at the company who hires me, I have a copy of the book The Design and Evolution of C+, by Bjarne Stroustrup. I brought it here because a colleague wanted to borrow it, but he never got around to doing it. I have actually read it myself, back when I tought C+ was cool.

Anyway, seeing the book lying there I remembered a thing from a couple of months ago, when I was working with what was supposed to become a book on test-driven development (which I’m not sure about continuing on now that Kent Beck is working with that).

I was thinking about whether you should cite or not and I remember talking with an academic friend of mine, and he said that some academic books cite a lot, while others just have an appendix with background bibliography at the back. I was, and still am, more in favor of the latter.

I also leafed through some of my computer books to see what was most common. I found two books that end up at the extremes of the citing spectrum. Firstly, Object-Oriented Analysis and Design by Grady Booch does A LOT of citing. It feels as the entire book goes like this: “As John Doe points out, contrary to the principles of Jeff Smith, blah blah blah—or, in the words of Jane Hoffa: ‘blah blah blah’.”

Secondly, Designing Object-Oriented Software by Rebecca Wirfs-Brock, et al, does NO citing at all. There’s not even a list of books at the end. If I remember correctly, and their only acknowledgement is that the CRC card technique that they build upon was invented by Ward Cunningham and Kent Beck.

My conclusion is that it’s okay to do no citing at all, but since I really appreciate lists of suggested further reading, I would include such an appendix in any book I write. Now it’s time for some serious test-driven programming.