## The History of Visualization

Visualization permits the mind to organize both small and large quantities of data in a way that permits it to gain insight and understanding about relationships in the data. There are many techniques with which to visualize data, just as there are many shapes and forms of data. Finding the techniques that bring a clear understanding of the data, while somewhat of a science, is still somewhat of an art.

Perhaps the earliest form of scientific visualization is taking a 3-D image and projecting it onto a 2-D surface in the form of a map [1]. The earliest known wall map of a village appears to date from about 6200 B.C., while the earliest known map of the stars appears to date from about 16,500 B.C.

While maps have been known from ancient times, statistical graphs are of a more recent era. William Playfair (1759-1823), a Scottish engineer, is the author of the pie chart, the bar chart, and the statistical line chart.

Looking at multiple variables and deriving useful information from them is difficult. Charles Joseph Minard (1781-1870) developed a format to show data tied to a timescale with a landscape background. In 1869, Minard produced two drawings. The first showed the march of Napoleon's army towards Moscow, starting with 422,000 men, and ending with 10,000 [2]. The second shows Hannibal's crossing of the Alps, starting with 97,000 men, and ending with 6,000, [3], thus tying together the bitter cost of human life in the pursuit of military glory.

Today, most people are familiar with scientific visualization as it applies to two areas. The most prominent of these is the weather map seen on their local channels. The second is in the field of medical visualization, particularly with 3-D ultrasound images of unborn babies.

## Overview of Visualization

### The Three Types of Visualization

The three types of visualization describe the extent to which contextual information needs to be included. Additionally, these types somewhat describe what the final visualization product will look like. The three types are:

• I See
• We See
• They See

In the real world, there is overlap between these categories, as can be seen below.

### I See

"I See" visualizations tend to be an interactive, exploratory visualization with a minimal amount of context added. Put another way, the focus is not per se on the visualization, but in helping researchers in their research. By visualizing the collected or simulated data, researchers can decide on the how to continue in their investigation. Perhaps the visualization shows errors that need to be corrected or additional data that needs to be collected. Thus, at this stage, the visualization is not very polished. An example of this visualization would be to use the open-source program ParaView for investigative purposes.

### We See

"We See" visualizations tend to be an interactive, exploratory visualization with a larger amount of context added. The focus is still not on the visualization, but on communicating the research to others in the same group or field. At this stage, the visualization is not very polished, but will contain more contextual information to help explain the situation. The point is to clearly present the information in an effective manner.

### They See

"They See" visualizations tend to be in presentations and contain a high degree of contextual information to help explain the situation to a lay audience. It is not an interactive, exploratory visualization at this point, but instead is a highly refined product.

### What is Scientific Visualization?

Scientific visualization is the mapping of data to images to gain an understanding or insight into the problem under study. It transforms the numeric into the geometric.

There are some pitfalls with any visualization to avoid. In particular, the 14 Ways to Say Nothing with Scientific Visualization is an enjoyable and educating read, if a bit dated. Better yet is to maintain a skepticism over the results of computational simulations and their derived visualizations. Nothing looks worse than when a physical explanation for a result is used when it was based on a computational mistake.

Back to Introduction to Scientific Visualization