| |
|
|
|
|
 |
NCSA NEWS |
|
|
|
|
|
Understanding the Nature of the Data
One necessary step in sonifying data is to understand
the dynamics of the researcher's work. This entails
examining their models, the dimensions of their
numerical simulations, and the changes in the data
occurring over time. Bargar adds "if the data isn't
time-based, we have to go back and add temporal
relationships. We have to first understand the form
of the data, then we need to determine how we
might interact with it -- do we passively view it as
we would a photograph or a movie, or can we
engage with the data, push or twist objects representing
the data, something more than flying through a
space? Then we examine what information the
researcher wants to learn from their data. Only then
can we create a portrait of what kinds of sounds
would fit those characteristics."
Unfortunately, this data doesn't always tell you what
it should all sound like. To overcome this, Bargar's
team creates some boundaries. "We develop production
rules first," he explains. "Then, along with the
researcher, we investigate what type of sound would
be appropriate, remembering that there may be no
precedent. If the data they're studying has a
naturally occurring aural relationship in the real
world it gives us a good starting point. Or possibly
there's an accepted framework of artificial or
mechanical sounds. We bring in our own ideas at this
point and, together with the researcher, we decide
what the data might actually sound like. Then we
connect it to some specific sounds."
As an example, Bargar points out some recent work in
sonifying numerically modeled storms with NCSA Senior
Research Scientist Robert Wilhelmson, who is also a
member of the Alliance's
environmental hydrology Application
Technology Team. "Our group
decided to provide what we thought were good
information rich and controllable listening
experiences based on Bob's data, and then we let him
decide what was most informative to his audience." In
this way the researcher can perhaps correlate the
sound to their previous knowledge and select the best
listening experience from their perspective.
|
|
|
|
|
|
