Database Reference
In-Depth Information
controls don't work.
The main drawback to the slide-show approach is that users can't explore the interface, although you
might be able to have some useful discussion based on the concepts they can see. But the lack of
interactivity will limit the amount and nature of the feedback you can get—people stop exploring after
the first few dead ends. You can only ask, "What do you think that would do?" a few times before the
enthusiasm tapers off. This kind of prototype may be sufficient to expose relatively obvious errors or
omissions in the design, but if you get vague positive feedback of the "Looks good!" variety, take it with
a large grain of salt. The From the Field box illustrates why I believe that low-depth prototypes are more
suited for demos or groups reviews than for usability testing.
From the Field: Usability Testing a Slide Show
"When I was new to the usability profession, I had a bad experience when I tried to conduct
usability tests using a series of linked HTML files. The intent was to ask users what they'd click
on and what they'd expect to see, and then proceed to the next page once they'd found the 'right'
link. But once the users caught on that there were only one or two live links per page, the
usability test quickly deteriorated into a game of 'find the hot spot.' The sessions were a complete
waste of time. It really damaged the credibility of usability testing at my company, and for
subsequent efforts I faced an uphill battle."
Anonymous
Paper Prototype
The look dimension can vary a lot for a paper prototype—hand-drawn Web pages would get a low
score, grayscale printouts of professionally designed screens would be medium, and color printouts
might even be high. The interaction factor has obvious differences—touching isn't exactly like clicking,
writing isn't typing, and the human element is artificial. Although working with a paper prototype can be
highly interactive, it's quite different from working with a computer, so the paper prototype gets a low
score here. But what's interesting about paper prototypes is their depth because people take up the
slack—the Computer decides what to do based on the user's inputs. Even if those inputs weren't what
was expected, the Computer can (to borrow my own words from the earlier discussion of depth)
gracefully prevent, detect, or explain any combination of choices that didn't make sense. So paper
prototypes often have good depth.
It is worth mentioning that paper prototypes support breadth only up to a point. Although in theory you
could create a 5000-page Web site out of paper, it wouldn't be practical. So if an interface (or more
precisely, a particular task for that interface) requires a large number of screens, it may become
overwhelming for the Computer. Although I've never counted, I estimate that most paper prototypes
I've worked on have had anywhere from a few dozen to a couple hundred screens, which provided
sufficient breadth and depth for the tasks the product team needed to test.
Note Although you might hear both the PowerPoint slide show and the paper prototype referred to
as "low-fidelity" prototypes, Table 12.1 reveals that they tend to have opposite characteristics.
This further illustrates why the term low-fidelity can be misleading.
DENIM
Interestingly, there are some computer-based sketching tools that let you draw screens by hand (using
a pen-and-tablet input device) and link them into functioning computer-based prototypes. One such tool
is DENIM, which can be downloaded from guir.berkeley.edu/projects/denim. The DENIM project is
headed by Dr. James Landay of the University of California at Berkeley. The idea behind DENIM is to
provide Web site designers with a unified approach to various tools, such as site maps, storyboards,
and individual page design. DENIM supports sketching input and a visual language—draw an X on top
of a widget, and it disappears. As of 2002, DENIM does not provide support for all the widgets and
interaction found in HTML forms and traditional graphic user interfaces, although enhancements are
planned. See Figure 12.2 for an example of a prototype created using DENIM.
Search WWH ::




Custom Search