Ola Fosheim Groestad (olag@ifi.uio.no) wrote 20/6/96:
> But I do think that allowing for many subjective illusions are
> desirable as long as that doesn't interfere too much with the
> inter-user communication. (like, "hey look at that ball over there",
> "what ball? that's a fish")
I, Chris Greenhalgh & Steve Benford had a paper in FIVE'95 (1)
discussing subjectivity and proposing a framework for describing
subjective environments. This met with a mixed response, so I thought
it might be interesting to discuss it here.
We define subjectivity as the ability to define viewer dependent
features in a virtual environment. A subjective environment is one
that intentionally supports subjectivity (ie I don't count
differences in views caused by network delays).
Multi-user VEs allow each user to view the environment from an
independantly controlled viewpoint and some also support Level of
Detail (LOD) operations but still users are seeing basically the same
environment.
We have proposed that more extreme deviations from objectivity should
be supported. One justifcation for this is that we believe there is
a tension between wanting to collaborate with others and wanting to
customise your view to suit yourself. Environments representing
abstract information are possibly more likely to benefit from
subjective features than those representing concrete information
(real world scenes, CAD models etc). This is because there is more
flexibility in the way abstract information can be displayed since it
has no "natural" 3D representation and therefore different users may
prefer different representations but still wish to collaborate to
some degree.
Obviously the more user's subjective views diverge the harder it will
be becomes to collaborate. For this reason its possible that in
practice users' views may converge and diverge depending on the
applications being used, the activities being performed and the
degree of cooperation.
While I was at SICS I worked with Kai-Mikael J-Aro to implement a
mechanism for representing users to each other in subjective
environments with different spatials frames (ie artifacts may be
located differently in each subjective view) and a subjective
environment for visualising information (papers on this work are in
preparation). The next step is to have some user trials and see how
these ideas work in practice.
Comments?
best wishes
Dave
1) "What You See is Not What I See: Subjectivity in Virtual
Environments", Dave Snowdon, Chris Greenhalgh and Steve Benford, in
Framework for Immersive Virtual Enviroments (FIVE'95) 18-19th
December, 1995, QMW University of London, UK.
ABSTRACT:
This paper discusses the issue of subjectivity in collaborative
virtual environments. First, we identify current uses of subjectivity
in virtual reality systems. We then examine three existing and
representative collaborative applications to identify potential
benefits of subjectivity. A framework for describing subjectivity
based on networks of shared data and transformations is introduced.
This framework is then used as a vehicle for discussing issues of
subjectivity, mutual awareness, representation and
embodiment. Finally, we touch on a number of further issues including
notions of evolving consensus views and some possible problems caused
by subjectivity.
Dave Snowdon
Communications Research Group Tel: +44 (0)115 9514226
Department of Computer Science Fax: +44 (0)115 9514254
The University of Nottingham E-mail: dns@cs.nott.ac.uk
Nottingham NG7 2RD, UK <http://www.crg.cs.nott.ac.uk/~dns/dave.html>
"This mind intentionally left blank"