Database Reference
In-Depth Information
Usability Heuristics:
These heuristics, as introduced and developed by Nielson and
Mack [1994], focus on the usability of the interface and are designed to be applied to
any application, thus are obviously of use to information visualizations. They will
help make sure that general usability issues are considered. These heuristics are dis-
tilled down to ten items - visibility of system status, match between system and real
world, personal control and freedom, consistency and standards, error prevention,
recognition rather than recall, flexibility and efficiency, aesthetic and minimalist de-
sign, errors handling, and help and documentation.
Collaboration Heuristics:
When interfaces are designed for collaboration, two addi-
tional major categories arise in importance: communication and coordination. Baker
et al. [4] developed a set of heuristics that explore these issues based on the Mechan-
ics of Collaboration [29]. As information visualizations start to be designed for col-
laborative purposes, both distributed [31, 78] and co-located [35], these heuristics will
also be important.
Information Visualization Heuristics:
While the usability heuristics apply to all
infovis software and the collaboration heuristics apply to the growing body of col-
laborative information visualizations, there are areas of an information visualization
that these at best gloss over. In response, the Information Visualization research com-
munity has proposed a variety of specific heuristics. Some pertain to given data do-
mains such as ambient displays [46] and multiple view visualizations [5]. Others
focus on a specific cognitive level, for instance knowledge and task [1], or task and
usability [66]. Tory and Möller [74] propose the use of heuristics based on both visu-
alization guidelines and usability. As explored by Zuk and Carpendale [84], we can
also consider developing heuristics based on the advice from respected experts such
as design advice collected from Tufte's writings [75, 76, 77], semiotic considerations
as expressed by Bertin [8] and/or research in cognitive and perceptual science as col-
lected by Ware [79]. Alternatively, we can start from information visualization basics
such as presentation, representation and interaction [68]. However, a concept such as
presentation cuts across design and perception, while representation advice, such as
what types of visuals might best represent what types of data, might be distilled from
the guidelines put forth by Bertin [8] and from an increasing body of cognitive sci-
ence as gathered in Ware [79]. Sorting out how to best condense these is a task in
itself [52, 85]. “At this stage of development of heuristics for information visualiza-
tion we have reached a similar problem as described by Nielson and Mack [54]. It is a
difficult problem to assess which list(s) are better for what reasons and under what
conditions. This leads to the challenges of developing an optimal list that comprises
the most important or common Information Visualization problems” (page 55, [85]).
Summary of Inspection Evaluation Methods:
While experience in the human com-
puter interaction communities and the growing body of information visualization
specific research indicates that heuristics may prove a valuable tool for improving the
quality of information visualizations, there is considerable research yet to be con-
ducted in the development of appropriate taxonomies and application processes for
heuristics in information visualization.
The currently recommended application approach for usability heuristics is that
evaluators apply the heuristics in a two pass method. The first pass is done to gain an
overview and second is used to asses in more detail each interface component with
Search WWH ::
Custom Search