The Value of Visualization

From CS294-10 Visualization Fa07

Jump to: navigation, search

Lecture on Dec 5, 2007



[edit] Readings

  • Visualization Research Challenges. NIH/NSF Research Report. Johnson et. al. (html) Read the PDF document.
  • The Value of Visualization. Van Wijk. (pdf)

[edit] Hazel Onsrud - Dec 09, 2007 03:41:17 pm

I really enjoyed Wijks' article and his discussion of visualization from a variety of angles as/for art, science, and technology. I thought it was fitting that he concluded a theme so common in this class, in the field, and, of course, our vision of the world in general: it's all dependent on your point of view.

[edit] Robin Held - Dec 09, 2007 11:12:55 pm

At first I was a little bit wary of van Wijk's apparently numerical analysis of visualization effectiveness. However, in his article, he shows how such a model can be very useful. His intention is not to attribute exact values to the number of users or learning curve of a visualization technique. Rather, the model provides a means to efficiently compare multiple types of visualization. For instance, technique A may have fewer users than technique B, but A is employed by its users several times a day, while B is used once a year, so the impact of A may be larger. van Wijk essentially uses a quantitative approach to provide a qualitative overall result, and it works very well.

[edit] N8agrin - Dec 15, 2007 12:52:39 am

I was also initially skeptical about Wijk's attempt to codify visualization, however his formula and methods are not ill conceived. As a means of visualization communicating economic value and knowledge transfer it may be a useful. Certainly I don't think that Wijk's formula should be used without thought, but using it as a frame of reference for discussing visualization's utility is worthwhile.

[edit] Ken-ichi - Dec 15, 2007 05:40:28 pm

I think most people who have ever discovered something looking at a chart realize the value of visualization. However, I doubt the reason advances from visualization research aren't applied has less to do with a lack of objective evaluation standards and more to do with poor dissemination and a lack of tools. I doubt most academics even realize that information visualization is an academic discipline, and would be unlikely to seek out research or specialists when attempting to visualize data. Most academics are also not programmers, and thus often don't have the technical means to produce some of the more advanced visualization techniques emerging from research. For example, many researchers deal with graph data one sort or another, but where is the simple graph visualization software package that produces compelling pictures out of the box (don't say Graphvis)?

[add comment]
Personal tools