Page 72 -

2.3 Conceptual models 41

There are a number of different kinds of conceptual models. These can be bro-
ken down into two main categories: those based on activities and those based on
objects.

2.3.1 Conceptual models based on activities

The most common types of activities that users are likely to be engaged in when in-
teracting with systems are:
1. instructing
2. conversing
3. manipulating and navigating
4. exploring and browsing
A first thing to note is that the various kinds of activity are not mutually exclusive,
as they can be carried out together. For example, it is possible for someone to give
instructions while conversing or navigate an environment while browsing. How-
ever, each has different properties and suggests different ways of being developed
at the interface. The first one is based on the idea of letting the user issue instruc-
tions to the system when performing tasks. This can be done in various interaction
styles: typing in commands, selecting options from menus in a windows environ-
ment or on a touch screen, speaking aloud commands, pressing buttons, or using a
combination of function keys. The second one is based on the user conversing with
the system as though talking to someone else. Users speak to the system or type in
questions to which the system replies via text or speech output. The third type is
based on allowing users to manipulate and navigate their way through an environ-
ment of virtual objects. It assumes that the virtual environment shares some of the
properties of the physical world, allowing users to use their knowledge of how
physical objects behave when interacting with virtual objects. The fourth kind is
based on the system providing information that is structured in such a way as to
allow users to find out or learn things, without having to formulate specific ques-
tions to the system.

A company is building a wireless information system to help tourists find their way around
an unfamiliar city. What would they need to find out in order to develop a conceptual
model?

Comment To begin, they would need to ask: what do tourists want? Typically, they want to find out
lots of things, such as how to get from A to B, where the post office is and where a good Chi-
nese restaurant is. They then need to consider how best to support the activity of requesting
information. Is it preferable to enable the tourists to ask questions of the system as if they
were having a conversation with another human being? Or would it be more appropriate to
allow them to ask questions as if giving instructions to a machine? Alternatively, would they
prefer a system that structures information in the form of lists, maps, and recommendations
that they could then explore at their leisure?