Programming as If People Mattered: Friendly Programs, Software Engineering and Other Noble Delusions [Paperback]

I liked one chapter so much that I typed out a softcopy myself by hand: a copy to read whenever I like. I mean Chapter 20 "The Ivory Tower". That chapter says ... We should reinvent the university experience of programmers to make the experience useful to the programmer. To become the first to teach programming well, let's try applying the approaches used to teach similar disciplines like architecture, or even anthropology, art, and drama.

Whereas "The Design of Everyday Things" has been updated and refined, this book is stuck in 1991. Many of the insights of the book are excellent, but there's a lot of material that is simply no longer relevant. I can only recommend this book to people who are willing to look past the pedantic style, occasional irrelevancies, and evaluations of decade old technology. I'd recommend Alan Cooper's book The Inmates are Running the Asylum instead, though that has its own problems.

If you can look past the obvious defects, there's a lot here for readers interested in user interface design. It's all anecdotal, but it's squares well with other quantitative works.

If you got this far in the review, there's a lot in the book to reward you for looking past the obvious defects. The primary source for Borenstein is his work on Andrew, a large Carnegie-Mellon University project, which, for various reasons, was reduced to a footnote in the history of computing.

One of the most notable observations a reader will make of the book is that the more things change, the more they stay the same. Many topics are just as relevant today as they were in 1991. For instance, his discussions on standards still offer insight. Part one starts us off on the basic problems of user interface design, and the chasm between regular users and programmers. Cooper offers a similar analysis, though the tone here is a lot more constructive.

The title of part 2, "The Dark Night of the Soul: The State of the Art in UI design" is a misnomer -- its real focus is the various people involved in UI design, including the HCI folks, programmers, and "the men in suits." Those familiar with the players can skip it. Much of this ground is covered in other books. If you always wondered why you can't get the cool toys from the MIT Media Lab articles, this explains why. The analysis of programmers is similar to Cooper, but with much less inflammatory tone. You can get the analysis of the management role anywhere - though his comments on them in the next section are excellent.

Part three is the meat. Borenstein gives us the benefit of his experience, with his "10 Commandments of user interface design." These are most excellent, and worth the price of admission. There's a few that feel like he might have been grasping a bit to reach the requisite 10, but that's a minor quibble. Most of the advice here is still valid, and you can easily think of modern software that could have benefited from this commentary.

Part 4 is everything else. Mostly, it contains introductory primers on development techniques, usability study, and project planning. Oddly, there's a strong chapter on the fundamental flaws in computer science education (circa 1991) that will ring true for many educated in that period.

I would love to see an updated version of this book. Much of Borenstein's advice still holds true, and a second edition could bring this wisdom to the development community. In the meantime, use the advice of part 3, and program as if people matter.