Geography Reference
In-Depth Information
example, if a resolution of two seconds between two (classified) values and a
mapping of ten different values are required; one ends up with a 20 s sound
display in the worst case. But as experience teaches, attention significantly
decreases after a few seconds so that a relative comparison of two (or more)
values released one after the other becomes very difficult. With that, duration is
not a suitable option for depicting quantitative data.
•
Rate of change
: The discriminability of different tempi depends on a variety of
factors, like the duration and the rest between sounds. Empirical values from
music state a relative discriminability of ca. 4 beats per minute; however, this
also requires a certain total duration. Like with the aforementioned parameter
duration an exact or reliable determination of absolute tempo values is not
possible for the human.
•
Number
: The number or repetition of a sound is variable that is not proposed by
Krygier (
1994
). But in analogy to the graphical method of symbol repetition also
the playback of a certain number of sounds is suited in order to represent
absolute values. However, depending on the actual number of the values (either
unclassified or classified) this playback can last quite a long time.
From these considerations we can conclude that
•
for the determination of
absolute values
only the use of respective number of
sound displays seems to be appropriate; however, as described above, this
method is strongly limited to the total amount of values;
•
for the representation of
classified values
with the aim of a relative comparison
pitch can be recommended as best choice: Pitch is very often denoted as
“brightness” of a tone, which very well describes the analogy to graphical
coding. In analogy to the graphical bi-polar coding (using two colors and varying
brightness) we can also apply two timbres in combination with varying pitches:
For example, a high flute sound represents an increase, whereas a lower tuba
sound stands for a decrease of values with respect to a certain zero, mean or
standard value (which might be encoded by an additional base sound).
With respect to
implementation issues
the general problem remains that sounds
have to be presented in a sequential and time-dependent manner, thus requiring an
additional
interactive release
for each use case. This can be realized through
pressing on loudspeaker symbols associated to objects (like county polygons) or
simply by mouse-over movements over those objects. In any case, an additional
explanation for this interaction is necessary.
Like with graphics, also acoustic coding requires a
suitable legend
. Again, the
sound legend needs a certain interaction concept which finally leads to a longer
processing time compared to the graphical counterpart and with that to a reduced
usability.
Finally, as a consequence of the sequential display, a simultaneous recognition
of values of more than one object, and with that a quick comparison “at a glance”
becomes impossible. This again reduces the effectiveness and efficiency of nearly
all comparison tasks.
