Database Reference
In-Depth Information
What Paper Prototypes Won't Find (Interaction Issues)
Paper prototypes are a rather blunt instrument when it comes to detecting many interaction problems.
Following are several types of problems I've seen in testing real interfaces that I doubt I would have
found with paper. This list is by no means complete—I'm willing to bet that there are many others.
Small Changes
On some interfaces, small areas of the screen change in response to the user's actions. For example,
some applications use the status line area at the bottom of the window to display messages, or an e-
commerce site might update a persistent shopping cart to show the number of items in it. These
changes are often subtle on a computer screen; with a paper prototype, it's much more obvious
because the Computer leans over the table and sticks something on the prototype. Thus, with any sort
of small or subtle change, a paper prototype can only tell you if users understand the change, not
whether they would notice it—you can't draw any conclusions about the latter.
Scrolling on Web Pages
On some Web pages, users don't scroll down, and thus they miss a link or content "below the fold"
that's relevant to what they're trying to do. You might try simulating the fold literally, by folding the Web
page where it would fall at a given resolution. If the user says he or she would scroll down, unfold the
paper so that the rest of the page can be seen. Another method is to use a cutout as shown in Chapter
4 . These methods may help you determine whether you've succeeded in placing the most important
information above the fold, but they're rough measures at best. Because a folded piece of paper is
more salient than the scroll bar on a Web page, if users "scroll" on the paper prototype it's risky to
assume that they also will on the real site. On the other hand, if users don't scroll with the prototype
and miss something important that's below the fold, then this perhaps does indicate a problem with the
page layout that you'll want to do something about.
Long Documents and Lists
Interactions with a long document or list are difficult to simulate adequately with a paper prototype. In
one usability study, I watched users navigate through a series of online documents containing industrial
regulations. As long as users clicked the down arrow they were fine, but the minute they moved the
scroll bar directly they'd end up in another chapter—given that there were hundreds or even thousands
of pages, even a small movement of the scroll bar landed the users several dozen pages away. This
problem was painfully clear from watching users use the real interface (and confirmed the team's
decision to limit scrolling to just the current document rather than the whole series), but we wouldn't
have seen it with a paper prototype.
Similarly, I've seen users have difficulty selecting the item they wanted from a long list. When a user
scrolls through a long list by clicking repeatedly, their attention is on the items in the list, not the down
(or up) arrow they're clicking. If the mouse cursor drifts off the scroll bar control, the user's next click will
land either within the list (selecting an item the user didn't want) or outside of it (causing the list to
disappear). Alternatively, the user might try to use the keyboard—it's common for users to type the
word they're looking for, such as "fr" to find France in a list of countries. Some interface elements (such
as the URL field in Internet Explorer) support this behavior, but drop-down lists don't. Many users aren't
aware of the difference and are surprised when typing the R whisks them off to Romania.
The "False Top" Problem
A false top is a Web page problem that's a combination of visual design and interaction. On some Web
sites, once the users scroll down, they are inhibited from scrolling back up all the way to the top
because at some point the page appears as though it's already there. (The scroll bar elevator does not
seem to be a sufficient clue that more of the page is out of sight above.) I call this a false top (see
Figure 12.3 ).A false top can cause usability problems (although usually minor ones) if the user has
trouble finding top-of-page navigation that's needed for the task.
Search WWH ::




Custom Search