background image
DigiCULT
.
Info
28
seeing data in his eyepiece, in this inter-
face the user navigates through the collec-
tion and sees the books. The main problems
are the overcharged screen and the diffi-
culty in seeing and accessing the collection
behind opened books. Another issue is to
maintain a maximum resolution for both
tools. Hence, two separated areas in the
same interface seems a good choice and,
to avoid switching jerkily from one to the
other, we considered two other integration
metaphors.
Horizontally structured workspace
The space is separated into two horizon-
tal zones (Figure 6). The user can switch
from one to the other by rotating around
the Y-axis. In the first zone, the collec-
tion is rendered and the user can move
the point of view. In the second one, the
books are always displayed on the ground.
Hence, when the user moves in the collec-
tion zone, the ground and books also move
behind to simulate their immobility. With
this metaphor, the user can also create sev-
eral reading environments depending on
their orientation within the reading area
and the view angle of the camera.
Vertically structured workspace
With the previous metaphor, after mov-
ing in the collection zone, some previ-
ously opened books in the reading zone
can be hidden. Moreover, half of the space
under the ground is not used. To avoid
these problems and limitations, we can
split the 3D space into two vertical zones.
Moving the point of view in the collection
zone will not affect the view in the reading
zone. To switch from one zone to the other,
a rotation around the Z-axis can be used.
In the reading zone, the user can still cre-
ate several reading environments by rotat-
ing around the Y-axis. Of course, the user
is immersed in the scene and will under-
stand navigation more intuitively than as
described here.
CONCLUSION FUTURE WORK
A
fter a design phase during which we
have studied the behaviours of the 3D
metaphors within our 3D interface,25 we
shall now evaluate our new prototype with
a panel of regular readers of the ABU and
CNum. We will also work on the annota-
tion interface in order to provide users with
the full service in 3D. With CNum this
third interface could be replaced by using a
standard word-processing tool. Within a 3D
environment it must be a specialised inter-
face in order to avoid switching context
between 3D and 2D interaction metaphors.
A solution would be to enclose 2D appli-
cations (e.g. word-processing software or an
HTML browser) within the 3D scene as we
have done in a previous project;26 howev-
er, today's situation is the opposite (render-
ing 3D scenes in 2D windows). We believe
that low-cost 3D graphic cards will lead
to full 3D interfaces. In such an environ-
ment, 3D models (our library for instance)
and 2D applications (word-processing tools
for example) could share the same common
interaction metaphors.
25See P. Cubaud, P. Stokowski and A. Topol (2002).
26See A. Topol, `Immersion of XWindow Applications into
a 3D Workbench' in Proceedings of ACM CHI'2000, The
Hague, Netherlands, April 2000.
Fig 4: The cockpit metaphor
Fig 5: Horizontally structured workspace
Fig 6: Vertically structured workspace

CNAM,

2004

CNAM,

2004

CNAM,

2004