Theoretical Foundations of Information Visualization - Information Visualization: Human-Centered Issues and Perspectives - page 61

(a) Routing
Visualization

(b) Visualization Transform Components

Fig. 3. A depiction of the revised network routing visualization transform. Nodes represent the
state of the data (e.g., a table of events) while edges represent operators or interactions (e.g.,
parsing the data). In this example, the network visualization is combined with a graph visualiza-
tion by embedding the results of the former within the latter. The graph itself is a composition,
merging a spanning tree and the original graph to layout the selected sub-graph. Back-
propagation of state due to interaction is included. The depiction is based upon an extended
Data State transform model.

4.5 The Value of Visualization Craft

A research program investigating scientific grounding for visualization is not meant to
diminish the importance of the engineering component of visualization. Visualization
is a tool for humans; engineering efforts form the basis of providing such tools. For-
mal exploration and design models can guide the creation of a visualization system;
however, as is the case now, multiple comparable techniques will often solve the same
problem. Thus, the craft of visualization - the confidence in design choices gained
through experience - will still be needed to decide between the choices a formal
model provides. Inspiration and creativity will not be eliminated by a more rigorous
foundation; the foundation will serve as a springboard for such endeavours.

4.6 What Is Left to Be Done

Formal foundations for a science of information visualization are still in a nascent
stage. Elements of complete exploration and transform models exist; however, they
have neither been reconciled with each other nor validated for their correctness. A
close collaboration between perceptual psychologists, cognitive scientists, visualiza-
tion researchers, and practitioners is needed to drive research into foundations: per-
ceptual and cognitive scientists provide the human-based foundations, practitioners
provide the case studies for observation and validation of models, and visualization
researchers will form the domain-specific bridge between them.
Humans and computers play an integrated role in the development and utilization
of visualization. A formal foundation would measure the efficiency of the former, and
guide the design of the latter in order to create a more effective whole.