Making a Paper Prototype - Paper Prototyping: The Fast and Easy Way to Define and Refine User Interfaces - page 67

Simulating Interaction
The paper prototyping motto is: "With a little imagination, you can simulate almost anything." There are
many aspects of human-computer interaction that a human being can simulate well enough that
usability testing provides useful feedback. But complex or subtle interaction usually can't be simulated
perfectly; as Chapter 12 discusses in detail, this is a drawback of paper prototyping.
Tooltips/mouseovers. Tell users at the start of the test that the real interface will have tooltips
(a.k.a. "those little yellow boxes that pop up") to explain the icons. Tell them that if they want to see
the tooltip for an icon, they can point to it and ask, "What is this?" and you'll tell them what the
tooltip would say.
Rollover/pop-up menus. These are conceptually similar to tooltips but are harder to simulate
orally because a whole menu pops up instead of just a word or three. On a computer, I've found
that what most users do is click to make the menu drop down, and then click again to make the
selection. This works well enough to show you what option the user would select, but it will also
mask some of the subtle problems that can occur with rollover menus.
Beeps. Simply say "beep" whenever the computer would, for example, when the user clicks
outside a modal dialog box.
Drag & drop. This interaction is a bit difficult to simulate perfectly. Keep in mind that many users
never even try to use drag & drop in an unfamiliar interface—they use the menus instead—so it
may not be something you need to worry about. But if drag & drop interaction is an integral part of
your interface, ask users to specify what they're dragging and where they're dropping it. The
Computer then talks through the visual changes that occur during this process. ("As soon as you
move into this area, the cursor changes to this, and when you release the mouse you see ...")
Right mouse menus. Tell users at the start of the test that right mouse menus exist, and if they
want to see one they should tell you that they're clicking the right mouse button, and then you'll
display the menu. As with drag & drop, don't be surprised if no one tries to use right mouse menus
in an unfamiliar interface.
Sliders, progress indicators. I don't usually bother to make widgets for progress indicators
because they can be simulated verbally, for example, by telling the user, "A progress indicator
comes up. It says 20% ... 60% ... done." (Or "1% ... 2% ...") If you need a slider for some other
purpose (such as a user input device), cut two slits in a piece of paper and use a strip of paper for
the slider.
Animation and video. Rapidly changing images can be hard to simulate. Sometimes it's easiest to
describe to the users what they'd see on the screen; other times a still picture (or a series of them)
will suffice. For short video clips, consider using a video player to show the video, as one product
team did when they were testing their multimedia Web site.
Web site links. When I first started paper prototyping Web sites, I'd take a high-lighter and
highlight everything that was clickable—text links, buttons, image maps, and so on. As you might
imagine, the pages looked pretty garish and all that highlighting proved to be more distracting than
useful. It was also extra work, so I dropped that idea. Now I tell users at the start of the test that
they if they're not sure whether something is a link or just a picture, they can point to it and ask, "Is
this a link?" and we'll tell them yes or no. (Or, "It wasn't originally, but apparently it should be.")
Scrolling. On some Web pages, there is important information "below the fold" and you're
interested in knowing whether users can find it. Although scrolling is a bit cumbersome to simulate
with a paper prototype, there are a couple of ways to do it. The first is to fold the paper so the user
initially sees only part of the page, and unfold it if they tell you they'd scroll down. This method is
good enough for gathering gross data about whether users scroll on a particular page or not. If you
need to do something fancier, make a cutout in a large piece of cardboard that's the size of the
monitor display. Put the page underneath and slide it up or down as the user "scrolls." ( Figure 4.17
shows an example.) But many development teams make a deliberate decision not to include
scrolling in their tests, leaving this question to later tests with the actual design.