You, and others, might find the excellent work being done at the Univ.
of Maryland's HCIL (Human Computer Interaction Lab) to be of interest.
Several of the projects I've reviewed deal with how people "drill down"
through large quantities of data to get at the information they seek.
http://www.cs.umd.edu/hcil/
(look at the "visualization" link, among others, in the upper left of
the navigation bar).
tom
maohai huang wrote:
Bill Hibbard <billh@xxxxxxxxxxxxx> wrote:
Hi Maohai,
There must be some way around this, because the user
cannot come close to distinguishing 1 million shapes
in a single screen. One thought is to plot the million
points as points, and to create another data object
selecting only points not near to other points and
draw them as shapes.
There are several issues with this approach. One is
that one has to assume what the user want to see and
how the user's system perform to decide how to do
pre-selection. And this is tricky for a general
system we are building. The second one is that the plotter will be used
as extendable and flexible building blocks for making
more complex applications. Pre-processing and pre-
selection could end up being too restrictive for yet
unforeseen useages of the plotter. An "honest" tool
that can be easily understood is a very good thing.
The default DataRenderers are very general
I have gone through the data renderer guide. It seem
hard to get working knowledge out of it. I worry about
requiring future maintainers of our plotter to be
competent in visad renderer to maintain our package.
but thanks for the suggestions. they are all options
that have pros and cons.
Maohai
---------------------------------
Do you Yahoo!?
Free online calendar with sync to Outlook(TM).
