Hochmair Hartwig
The Wayfinding Metaphor— Comparing the Semantics of Wayfinding in the Physical World and the WWW
Abstract:
Wayfinding
is a common human task. The terms ‘wayfinding’ and ‘navigation’ are
traditionally associated with an activity that takes place in the real
world. The development of new electronic media induces humans to
navigate artificially created environments, e.g., the World Wide Web
(WWW), computer games, or virtual environments. Although real
environment and artificial environment show different features—e.g., in
the definition of a distance between places or in the organization of
space—we claim that the concepts of wayfinding in the real world can
also be found in the WWW.
A goal of the thesis is to determine
what the term wayfinding means, i.e., to describe the semantics of
wayfinding. Analyzing several wayfinding definitions in literature we
found that there is no unique meaning for the term wayfinding, although
there seem to be some core properties of the underlying process.
Therefore we consider wayfinding to represent a radial category. From
the analyzed definitions we get the central meaning of wayfinding, and
describe it through a set of axioms. The axioms define constraints on
agent and environment. If the axioms are satisfied, the activity
performed by the agent describes a wayfinding process.
Another
goal of the thesis is to show that within the wayfinding metaphor, the
semantics of wayfinding is similar for both the real world and the WWW.
We hereby abstract the conceptual wayfinding model through algebraic
specifications and give two parallel instantiations. We show that both
instantiations satisfy the axioms, and thus the term ‘wayfinding’ can
also be used for the Web space—expressing a similar semantics as in the
physical world.
The axioms are invariant under the applied
strategy and the type of environment. Therefore we can choose any
wayfinding strategy that is capable of coping the wayfinding tasks
given in the two cases studies (where the environment is unknown to the
agent). The chosen wayfinding strategy relies on ‘information in the
world’ and applies a semantic decision criterion. A wayfinding
simulation shows that the formal algebraic specifications of the
agentbased model are executable.
download (in pdf format)

