SimpleWeblog

Book Review: The Inmates are Running the Asylum (Alan Cooper)

This is a book where Alan Cooper (author of HCI classics like About Face and "the father of Visual Basic") talks about why interaction design (what he and his company do) is so important. It is supposedly directed at a more business audience, but also provides a nice introduction to a lot of his philosophy on why software products suck so much.

And I totally agree with him that most software sucks.

His 2 main points are:

1. You need to design software before you start programming it.

2. Programmers shouldn't design the software.

Which, in general, I think I agree with. Yay. But I had some issues. I wonder if a lot of my problems resulted from him overgeneralizing for his "business audience" to make sure they understood his points. But he goes on and on about how programmers are not like "normal" people. And that is true, sort of, but that doesn't mean all programmers are alike, either. Maybe I just take it personally because I don't feel like I fit into his category of programmers who all like to be in control of everything technological.

Maybe I'm lazy, but I like to be able to concentrate on what is important to me, and not on installing drivers or configuring networking parameters or resolving conflicting anything.

And it seems weird that he goes on and on about how programmers don't make good designers even though he was (is?) a programmer. He does try to refine the point a bit, that you can't do both at once, and it is better to have people who won't worry about programming concerns doing the design so they can focus on users more, but it struck me as a little bit of "I have made the switch, but most programmers can't." Which, again, may be because of his audience for the book. Or could be me being touchy.

His "goal oriented design" approach seemed interesting, though. You define personas, which represent specific people who will use the software. Then you create the goals for the personas. Then you create scenarios describing typical interactions with the system for the personas. Then, I guess, you actually design the system for the personas.

And there I think I was turned off by his dismissal of a lot of the more "rigorous" user testing that is typical talk for HCI types. He pulled a "rely on our experience" line, which makes me think of all of the psychology research showing that drawing on experience is not as useful as you'd think in most fields compared to some simple, "rigorous" analysis.

A lot of the book also comes across as a commercial for his company. Which is ok, I guess, since the company is a perfect example of what he wants the process to work like, but still strikes me as a little odd.

If it had been about 50 pages shorter, it would have been much better. But it was interesting, and did give me some good ideas about "interaction design." And it all made me feel a little better about our current process.

Posted by babar at April 9, 2001 10:15 PM