What Paper is (and Isn't) Good For - Paper Prototyping: The Fast and Easy Way to Define and Refine User Interfaces - page 199

Figure 12.2: Computer-based sketching tools like DENIM allows you to draw using a pen input
device and then link pages together to make a "hand drawn" prototype, which can then be tested
on a computer.
Thus, a DENIM prototype has a deliberately rough look because its goal is to support sketching. I
would rate its interaction capability as medium and its depth as low to medium because it doesn't fully
support all types of interface widgets, but these scores might change as more functionality is added to
DENIM.
When it comes to usability testing, a tool like DENIM may not offer any compelling benefits over a
paper prototype other than removing the human Computer from the equation. And that may not even
be its intent—much of DENIM's functionality is aimed toward providing a productive environment for
designing, not necessarily for usability testing. To me, one of the intriguing aspects of a tool like DENIM
is that it may enable researchers to tease apart the effects due to the medium of the prototype (paper
or computer) and the completeness of the design.
Methods Don't Stand Alone—What Are You Prototyping?
Table 12.1 was created in the context of an e-commerce site, so it may or may not be relevant to what
you're doing. The prototyping methods alone don't completely determine the score for each
dimension—it depends on what you're prototyping. For example, you could make an HTML prototype of
a handheld device as shown in Figure 12.1 and its interaction would be only medium because clicking a
button with a mouse isn't the same as poking it with your finger. So if you're prototyping a handheld
device, you might conclude that HTML prototypes get a medium score for interaction. On the other
hand, if you're building a Web site, an HTML prototype would get a high score for interaction.
Here's an ironic example: You'd think that a paper prototype would get high marks in interaction if
you were prototyping a touch screen device, but this isn't the case. Touch screens are
complicated by factors such of calibration and parallax, which affect how the computer and
human interpret each other's behavior. In practice, there are some important subtleties about
touch screens that you can't adequately simulate with a paper prototype. Funny, huh?
So what are you prototyping? If you're feeling ambitious, you might make your own version of Table
12.1 for your interface and the prototyping methods you're considering.
[ 1 ] Strangely enough, a few years after he started using the term fidelity, Tom Tullis went to work at
Fidelity Investments. Coincidence? Clairvoyance? Or even... conspiracy?
[ 2 ] If you read my descriptions of breadth and depth and wonder, "Isn't she talking about horizontal and
vertical prototypes?" the answer is essentially yes. The terms horizontal and vertical are common in
prototyping literature—I chose to use breadth and depth simply because it's more expedient to use
nouns for the dimensions instead of adjectives.
[ 3 ] I hope to be safely retired from the usability business before I'm asked to test an interface
incorporating smell and taste!