Once there is a data description that DX can read, we proceed to the visualization programming stage. This can be described simply: input, processing, output, just like all other computer operations. Generally, one of the first operations in a net is Import to fetch a data file from a file system into the memory controlled by DX. Then a series of operations are performed and an output, usually an Image or Display module presents a visual representation to the viewer.

Now you know 2/3 of any DX program. Import and Image. All we have to do is fill in that middle part: small matter of programming. Only ~161 modules to go! (though we really won't be describing or using every single one of them). It's in the middle processing stage where we get to investigate different kinds of visual representations or realizations and tune the imagery for output to various media such as video (made of a series of images), stills for journals or the web, or programs designed for interactive investigation by yourself or your colleagues using OpenDX on this and future similar data sets.

To do that, we'll need to learn the physical manipulations required by this graphic user interface to the visual programming language, then we'll explore in more depth what particular modules do. Each module manipulates DX Objects, like Fields, Groups, and so on, in a particular way. Some modules output slightly modified Fields or Groups; others select Components or even individual values from one Object and insert them into other Objects. Some take a certain kind of input Object and output a very different kind of Object. Learning what each tool does is like learning any new tool set: once you know what operation is called for, you'll be able to reach for the correct tool and know how to apply it.