Database Reference
In-Depth Information
mechanisms to highlight individual data items across representations might aid
individuals when switching between group and parallel data exploration.
Findings suggest that the availability of multiple, interactively accessible
representations might be important for information visualization applications
since the availability of multiple data representation can change decision mak-
ing strategies [52]. Also differing representations have an influence on validation
processes in information analysis [79], and more easily support people working
in parallel on information tasks [72]. While this is probably applicable, empirical
evidence directly linking these finding to collaborative information visualization
has not yet been gathered.
It is also possible that the actual mappings used in representations may have
to be re-thought. For example, spatiality or the use of position/location is com-
monly an important aspect of representation semantics. However, spatiality as
manifested in territoriality is a significant factor for communication and coordi-
nation of small group collaboration. It is an open question as to whether there
is a trade-off between these two uses of spatiality.
Presentation Issues:
Presentation has been defined as 'something set forth for
the attention of the mind' [63] and as 'the way in which suitably encoded data
is laid out within available display space and time' [87]. From these definition
is clear that changing display configurations, as is usually the case to support
co-located collaboration, will impact the types of presentations techniques that
are possible and/or appropriate. Common presentation techniques include pan
& zoom, focus & context, overview & detail, filtering, scrolling, clutter reduc-
tion, etc.
A common theme in information visualization is the development of pre-
sentation techniques that overcome the problem of limited display space (e. g.
[4,20,49]). In collaborative scenarios, information visualizations might have to
cover larger areas than in a single user scenario as group members might prefer
to work in a socially acceptable distance from each other. The display space
might also have to be big enough to display several copies of one representation
if team members want to work in parallel.
If groups are working over a shared presentation of data, presentations might
have to be adapted to allow collaborators to drill down and explore different parts
of the data in parallel. Collaborative information visualizations will likely have
to sup-port multiple simultaneous state changes. This poses additional problems
of information context. Team members might want to explore different parts of
a dataset and place different foci if the dataset is large and parts of the display
have to be filtered out. Information presentations might have to be changed to
allow for multi-focus exploration that does not interfere with the needs of more
than one collaborator. For example, DOI Trees [18] or hyperbolic trees [55] are
examples of tree visualizations in which only one focus on the visualization is
currently possible. ArcTrees [67] and TreeJuxtaposer [65], for example, allow
for multi foci over one tree display but these were not designed to take the
information needs of multiple collaborators into account and might still occlude
valuable information.
Search WWH ::
Custom Search