Visualization over the Internet Andreas Kolb ak@teleweb.at Institute of Computer Graphics Vienna University of Technology Austria

1. Introduction

1.1 WWW as a platform

One of the most impressing developments in communication technology during the past decade has been the amazingly fast evolution of the World Wide Web based on the Internet. The useful combination of intuitive web browsers with similar look and methods of interaction on multiple platforms and servers spreading information all over the web provides a powerful facility to distribute information to a large number of users. Information sharing was never easier than now with the facility Internet.

The main advantage of the Internet is it’s independence of hardware and software and thereafter makes it possible to connect new developments in the field of scientific visualization and almost any other topic with a large community of users.

1.2 Motivation

Scientific visualization is typically used to illustrate and investigate numerical data with the aim on images that are easier to interpret than the raw underlying numbers. The development of new visualization techniques is usually done by scientists using expensive specialized hardware and visualization software which is in common ways not accessible to the ordinary users. The results of their research can only be passed to other scientists with similar hard- and software or some lucky users who are also owner of such special devices and software components. This is usually only a small subset of potential users, but to reach a broader field of audience, the developed visualization techniques have to be implemented on other, common platforms as well. This step is very time and money intense and, of course, not taken by scientists at all. Because the number of users who have a desktop PC is significantly higher than the number of users currently working with high performance visualization systems, a visualization solution running on simple desktop PC’s will improve the distribution of innovative methods from scientific developers to users significantly. The benefits for the developers would be more feedback from a broader field of users and thereafter the possibility to tune the system so that it fits the users best.

To cope with that problem the world wide web might be the key. Due to its "near" platform independence and intuitive way to use it hides the complexity of the beneath internet and connects computers from various types and platforms.

1.3 Demands on a visualization technique

To be broadly accepted by the user community a good visualization technique must cope with the following problems and include these important features:

• Portability: The visualization technique should be portable and run on all kind of common platforms to grant a great amount on potential users the usage. Of course, the most common platforms like UNIX workstations, desktop PC’s using WindowsNT and Macintosh computers should be covered in addition to special scientific hardware. Java and VRML both provide the power to realize this.
  
• Intuitive usage: The technique offers a set of predefined visualization configurations covering common problems and presents the same intuitive interface on each platform to ease the use for people with more than one computer and/or when changing the operating system. Almost any browser released up to now meets this requirement. The user shall also be able to use at least one sample data set to test the visualization technique if it meets his requirements.
  
• Flexibility: The visualization technique should be flexible, allowing the user to visualize the data in a customized way with a minimum amount of work due to premade configuration settings. The user shall also be able to go into further details as he is given the possibility to tune all the parameters of the premade configuration by hand. It should also be of no great afford for the user to visualize his own data rather than given sample data sets.
  
• Interactivity: The illustration and the investigation of the data set through the user shall be an interactive process. The user should be able to change parameters, points of views and set filters while investigating the data to make the visualization a living process and to invite to "play with the data".
  
• Unambiguousity: The technique shall be at least rudimentary safe to misinterpretation of the visualized data due to wrong settings or parameters. This is a feature who bears much responsibility because a visualization technique with to much overwhelming parameters to change can be as useless as a technique with to much restrictions.