OpenDX.orgThe Open Source Software Project Based on IBM's Visualization Data Explorer
Home Page
Recent News
About OpenDX
Getting Started
The DX Community
Contact Info


All rights reserved.

Latest News

If you're using OpenDX and have some nifty pictures you'd like to share, contact us to get them added to the gallery.

  • October 14, 2007 - The CVS is again available. After being off-line at IBM for 4 months, it is now back up and an OpenSource repository. Check it out in the Developers section.
  • September 11, 2006 - The new OpenDX DVD is available for purchase from VIS, Inc.
  • September 9, 2006 - New mailing lists are up for use. Please come join us there if you prefer e-mail over user forums.
  • August 30, 2006 - OpenDX 4.4.4 is now available. Binaries for the highest demand operating systems are now available. See the ChangeLog for the list of changes.
  • March 22, 2005 - Jerry Hagon has updated his font converter script with some helpful documentation changes. Click here to download it. He also has a new dvidx which allows dvi output to be imported into OpenDX. Check out DXfontutils for more info.
  • July 1, 2004 - VIS, Inc. has updated its book, OpenDX: Paths to Visualization, to version 2 reflecting the changes in OpenDX 4.3.

User Quote

"The time required to assemble, test, and utilize a new application using IBM DX is phenomenally small. This characteristic is primarily due to the well designed data model, robust surrounding software, clean user interface, and dedicated IBM research staff."

T. Todd Elvins, January, 1999
San Diego Supercomputer Center

Data Explorer in Action

A basic tool for examining relationships between variables is a scatter plot. The plot shown here is from a program that lets the user interactively select which variables will be plotted on each of the axes from a list of fifteen variables in the original file. The variables are both numerical (e.g. hours per week and age) or categorical (e.g. gender). In the input file categorical values are strings. The program assigns a numerical location to display the string variables, and automatically puts the correct text labels (e.g. Male, Female) on the scatter plot. Different glyphs (symbols) can be used for the data points -- here 3d crosses are used because they are relatively fast to display, but easier to see than individual pixels. The data values are color coded to display a fourth variable. In this type of plot, this is particularly effective when the variable takes on a small number of distinct values (in this case two -- salary above $50,000 and salary below $50,000) that can be assigned distinct color hues (in this case red and blue).