General Question of Task/Network Analysis
Zack Jacobson: How does this help us envision visualisation?
The temporal aspect seems to be lacking. Iād like to have a better perception of what changes in time are meaningful within the psychology of visualisation, and what impact visualisation is going to have on a person.
That is one of the things we have been investigating. We are sensitive to certain time frames, it is no good to be too slow or too fast. Humans are more or less sensitive to movement in certain colour schemes. Also, we want to display information differently in different time scales.
In response to the question about whether there any guides to what we create and what they are going to look like, I think we should make things look as natural as possible. Things should change in ways that seem natural. It is better in terms of understanding.
There is a problem with computer memory.
But we should aim for that.
Colin made a significant comment with respect to when you add motion, you have to worry about peopleās problems with speed. There is also the issue of persistence of the image in the mind. You have to be careful that when you delay the image as the user may forget what you are trying to contrast the new image with. The issue is the time gap.
The answer is in time mapping techniques.
What about mapping it against colour?
The time factor must be spatially static, or else you build what appears to be as solid. You know that there is space in there and that there are things that you want to see, but it appears to be solid.
The minute you add annotations you cover something, and you forget what is underneath them. You need to have the annotation float so that you can push it out of the way.
That is easy enough, but is still a problem.
Colin, Iām curious about the use of stereo.
We donāt use stereo, as we decided that it wasnāt worth it.
Can you can rock the scene?
Yes, you can rotate it.
You are going to have to be careful with depth perception.
Iāve seen his stuff, and that is not a problem.
We use stereo with other stuff, but it becomes very challenging with these types of diagrams. Texture really enhances stereo.
Can you imagine using this technique to show the relations of things on the battlefield?
A lot of the same types of paradigms are occurring, for instance in data drill down and in the ability to query.
There is a dynamic aspect to battlefield icons, they change in location and status over time.
What about fields of influence of artillery? The icon maps could show the field of a possible manoeuvre, the terrain and the change in potential strategies. Both kinds of approaches are suitable to battlefield problems.
If it were more automated, then the table maps would lend themselves more to that use.
Spatial reasoning has a far broader meaning then just the relationship between different shapes. It means things that involve distance, many objects, geometries, etc.