not paying attention to the whole process, but rather just the end result Is that so wrong when the "whole process" is frequently opaque, confusing, error-pron

not paying attention to the whole process, but rather just the end result Is that so wrong when the "whole process" is frequently opaque, confusing, error-prone, poorly explained, and irrelevant to the task at hand? Be aware that you're subject to a cognitive bias: You find computers inherently interesting and worthy of attention. (You wouldn't be here if you didn't.) Other people, perhaps people whose initial experiences with computers are unlike yours, do not feel the same way and do not have the same intuitive sense that the "whole process" is worth attending to that you do. That is to say you have something like hindsight--it's obvious to you, bearer of a certain understanding, but to someone who doesn't know what you know, it isn't obvious what things they should be paying more attention to or what specific things they need to understand. Too often we think "They should know more about computers", which is very vague and ignores all of the things that people do tend to learn quite well. Your understanding comes from years of built-up cognitive models about how these things work. The thing is: So does theirs. Like you, they've built a model that works well enough to do what they want to do. The difference is that they've learned things like "If I meddle with things I don't understand, I might break something." and "If I break this computer, I won't be able to fix it." and "If I can't fix it, I will feel stupid and have to ask or pay for help." These are maladaptive (though too often true), but they are still learned things, and I think it's a big mistake to ignore them when thinking about why users seem to be unwilling to learn other things. ...tools that are actually designed to make their experience easier, faster, and more efficient. Those features and tools are not perfect. They fail. But form follows failure, which is why for instance several browsers now default to an interface that shows screenshots of frequently-used sites (the others all default to search pages). This accomplishes the approximately same thing as bookmarking or setting a home page without user action at all, and without relying on page titles or URLs for recognition. In time, the ways that this fails will become evident, too. Bear in mind, too, that unless they have some specific motivation or inspiration for change, people are more prone to do what they know works than to experiment for the sake of it. The RWW story is particularly relevant because of the often ignored detail that what those users were trying to do (Google for "facebook login") actually works most of the time. Until a series of basically reasonable assumptions failed, they had no intrinsic reason to do things differently.