Take any job and separate the workers into quartiles according to how much they produce. For any job, the output of the workers in the top quartile will be about double that for workers in the lowest quartile. The factor in software engineering is not 100%, though. It's more like 2000%. Amazing huh? That's right...the top software producers can do the same job in about 1/20th the time as the slowest ones....and their product is usually better. Contrast this with the way software is done in many large software factories where there the managers scoop up an armful of programmers and toss them at a problem.
This book is required reading where I work. I wasn't resentful so much as skeptical at first. I've seen a lot of important concepts come and be superceded over the years: structured programming, top-down design, object-oriented programming, api (everybody has them now!), object-oriented design, client-server, user-friendly etc.
It's not so much that these are bad ideas, or even that they're fads, so much as the fact that people who don't really understand them very well are using them as if they did understand them and are really making a mess of things in the process - first by messing up projects and next by passing on bad practices to other engineers. The worst offenders are the briefing slide gurus...you know the type.
But this idea of usability engineering is a good one - just like the others above. I only hope it becomes an accepted part of engineering practice and more than just buzz words on a briefing slide.
There are some case studies backing up what Nielsen is saying and some good explanatory remarks. He also explains what usability engineering is and how it fits into the scheme of system acceptability. He then goes onto to give some specific examples of things that can be down to implement simple usability engineering techniques. As it turns out they are some of the same things I learned in college software engineering courses years ago. But the questions is: why are these things (some of which are already being taught in universities) not in wider practice? Part of the problem is with the briefing slide gurus who think that the mere mention of a word makes it so. "Hey, we just put the words USER FRIENDLY on the box and that's magically what it becomes!" Turns out that the briefing slide gurus like the buzzwords, but they don't have any grasp that efforts like usability (and the others) have to be on the schedule, in the budget, and fully staffed.
But enough of my rants. Nielsen says this directly and without the polemics. But that's the point. The best stuff in the book though is the good ideas and advice filling the pages. Examples: Page 194 has quantifiable usability measures. Pages 181-184 discuss the ethical considerations for usability testing. Chapter 5 discusses usability heuristics. EXTENSIVE BIBLIOGRAPHY in appendix B. Also discusses usability life cycle, international interfaces, and assessment methods.
I read it in one day (maybe two..i forgot). I'll probably reread this in a few months. Worth every second for those of us in the computer software field.
