Page 211 -

I 180 Chapter 6 The process of interaction design

power outlets will be dependent on how the wiring within the building is designed
and the capacity of the main power supply; the choice of materials used in a pho-
tocopier may depend on its friction rating and how much it deforms under certain
conditions.
In an interactive product there are similar factors that are externally visible
and measurable and those that are hidden from the users' view. For example, ex-
actly why the response time for a query to a database (or a web page) is, say, 4 sec-
onds will almost certainly depend on technical decisions made when the database
was constructed, but from the users' viewpoint the important observation is the fact
that it does take 4 seconds to respond.
In interaction design, the way in which the users interact with the product is
considered the driving force behind the design and so we concentrate on the exter-
nally visible and measurable behavior. Detailed internal workings are important
only to the extent that they affect the external behavior. This does,not mean that
design decisions concerning a system's internal behavior are any less important:
however, the tasks that the user will perform should influence design decisions no
less than technical issues.
So, one answer to the question posed above is that we choose between alterna-
tive designs by letting users and stakeholders interact with them and by discussing
their experiences, preferences and suggestions for improvement. This is fundamen-
tal to a user-centered approach to development. This in turn means that the de-
signs must be available in a form that can be reasonably evaluated with users, not
in technical jargon or notation that seems impenetrable to them.
One form traditionally used for communicating a design is documentation, e.g.,
a description of how something will work or a diagram showing its components.
The trouble is that a static description cannot capture the dynamics of behavior,
and for an interaction device we need to communicate to the users what it will be
like to actually operate it.
In many design disciplines, prototyping is used to overcome potential client
misunderstandings and to test the technical feasibility of a suggested design and its
production. Prototyping involves producing a limited version of the product with
the purpose of answering specific questions about the design's feasibility or appro-
priateness. Prototypes give a better impression of the user experience than simple
descriptions can ever do, and there are different kinds of prototyping that are suit-
able for different stages of development and for eliciting different kinds of infor-
mation. One experience illustrating the benefits of prototyping is described in Box
6.2. So one important aspect of choosing among alternatives is that prototypes
should be built and evaluated by users. We'll revisit the issue of prototyping in
Chapter 8.
Another basis on which to choose between alternatives is "quality," but this
requires a clear understanding of what "quality" means. People's views of what is
a quality product vary, and we don't always write it down. Whenever we use any-
thing we have some notion of the level of quality we are expecting, wanting, or
needing. Whether this level of quality is expressed formally or informally does not
matter. The point is that it exists and we use it consciously or subconsciously to
evaluate alternative items. For example, if you have to wait too long to download