Stability: Designers focus on system "stability," but it's not because they worry about time wasted, though that's how users tend to talk a lack of stability. Like the performance issue, instability is about the theft of system control from the user. People waste all kinds of time on all kinds of things, and usually don't mind doing it. What enrages people is when somebody else forces wasted time on you. Blue Screens of Death are more akin to running into unexpected traffic jams or having somebody take away the TV remote control. You're forced into putting your objectives on hold, and left feeling powerless.
One reason for the industry-wide pandemic of frustrating products is that the whole culture of usability testing doesn't emphasize user feelings of control. Microsoft does usability tests, for example, but its tests are flawed. Typically, it sits random people in front of a PC in a usability lab. Victims are directed to do various tasks, and asked what they're doing and thinking as they try to complete those tasks. All of this is monitored, and everything is recorded.
Microsoft usability testing tends to focus on enabling users to "accomplish goals." Microsoft categorizes these goals according to their educated preconceptions about what people are trying to do based on their jobs or user categorization are you a student, middle manager, designer, for instance. So Microsoft focuses on results. My view is that how the user feels during the process is more important than anything else.
Here's the problem. In these scenarios, users are using somebody else's PC. They expect and assume that the software is in control. There is no psychological feeling of "ownership" over the equipment or the software or the work or anything. So the most important element -- the sense of control people feel when doing their own work on their own PCs in their own homes -- is missing entirely from the tests.
During usability tests, users are asked constantly about the software. And that's the wrong question. When real people are doing real work, they're focused on their own desires and objectives and are frustrated or not frustrated based on the degree to which they're given what they want.
My advice to Microsoft is to add an additional test: a "Who's In Control?" test. After performing a task, ask the user to rank their experience on a scale with "me in control" on one side, and "software in control" on the other. Try all test methods for completing various tasks, and choose the one ranked with the maximum "me in control" score. And they need the home version for ongoing testing in the "real world."
We've all experienced the full range of emotions while using gadgets, PCs, phones and software. At one end of the spectrum is a kind of thrilling joy, where something "just works." At the other end, there is a consuming rage. The amount of time your emotional state spends at one end of the spectrum rather than the other is the one and only thing that determines how much you "love" the product.
All the factors involved in using a PC -- consistency, usability, simplicity, stability, performance and even the successful completion of tasks -- all come down to control.
Give me control, and I will love your product. It's as simple as that.
Mike Elgan writes about technology and global tech culture. Contact Mike at firstname.lastname@example.org.