Database Reference
In-Depth Information
The data centric approach discusses factors such as who has (or does not have)
rights to which parts of the data?, who can change the scale, zoom, or rotation
settings for a shared view of the data? And how does a data item get passed
between team members (hand-off). Restriction has been suggested as a means to
stop certain members from making unsuspected global changes to the data that
might change other members' view of the same data [75]. Similar issues pertain-
ing to workspace awareness (individual vs. shared views), artefact manipulation
(who can make which changes), and view representation have been raised [39].
Is a single shared representation adequate? Should a system allow for multiple
representations? Should the exploration on multiple representations of the same
dataset be linked or be completely independent?
Fluid Interaction:
The fluidity of interactions in a shared workspace influences
how much collaborators can focus on their task rather than on the manipulation
of interface items [82]. This implies that in a collaborative information analysis
scenario, parameter changes to the presentation or representation of a dataset
should require manipulation of as few interface widgets (menus, slider, etc.) as
possible and little or no changes of input modalities (mouse, keyboard, pen,
etc.). A study on collaborative information visualization systems has similarly
reported that groups worked more effectively with a system in which the required
interactions were easier to understand [61]. This poses a challenge to information
visualization tool designers as typically a high number of parameters are required
in visualization systems to adapt to the variability in dataset complexity, size,
and user tasks.
4 Collaborative Visualization on the Web
Visual analysis is rarely a solitary activity. A business analyst may notice an
unexpected trend in a chart of sales figures - but then she's likely to confer with
a colleague, who may share the chart with a manager, who later might present
it to executives. Such scenarios of collaboration and presentation across both
time and space are common in business and scientific visualization. Just as a
good visualization takes advantage of the power of the human visual system,
it can also exploit our natural social abilities. Accordingly, designers of visu-
alization systems should consider not only the space of visual encodings, but
mechanisms for sharing and collaboration. At minimum, systems should enable
people to communicate about what they see so they can point out discoveries,
share knowledge, and discuss hypotheses.
The social aspects of visualization have taken on new importance with the
rise of the Web. While collaboration in small groups remains ubiquitous, it is
now also possible for thousands of people to analyze and discuss visualizations
together. These scenarios are driven by the fact that users can interact remotely
from anywhere on the globe and access the system at different times. Parti-
tioning work across time and space holds the potential for greater scalability
of group-oriented analysis. For example, one decision making study found that
Search WWH ::
Custom Search