At this point, you should turn to the online documents and read User's Guide Chapter 2, then Appendices A.2-A.5 for more background on Data Explorer. Skip Appendix A.1 until later, as this is a bit technical if you are still new to Data Explorer.

The PDF files are found at: http://opendx.org/support.html

Or, you can navigate directly to the Users Guide: http://opendx.npaci.edu/docs/pdf/userguide.pdf

Footnotes to Chapter 2, Appendices A.2-A.5

These footnotes are too technical to understand until you've done some of the exercises in this workshop, but I needed to insert them somewhere. Just make a note to yourself to read them later.

On p. 234 (Appendix A.3), I make the remark that you can tell Switch to send nothing from its output by feeding it the value "0". While this is true of Switch, you may get errors from downstream modules that expect a valid object: Switch = 0 actually sends out a NULL object which is unacceptable to many other modules.

The last paragraph of the Switch discussion on p. 234 suggests using Selector (on/off) and Switch, but this is not the best solution any more. You are better off using Route, another module used in flow-control; the Route module did not exist when I wrote this material in 1993. By feeding Route a "0", you effectively turn off all processing below Route, and therefore get no errors from downstream modules.

I routinely add Selector-Route pairs to my nets that together make handy On-Off switches (since the default Selector is already set up to emit 0 for Off and 1 for On). This way, you can turn on or off any portions of your net while you are developing a new portion, or debugging a section (turn off all the parts that you know work, to narrow down the problematic section).

On p. 234, in the discussion of Transmitters and Receivers, I do not mention another very important use of these modules, because it had not been invented in 1993: Pages. To send information across pages in your network program, you must use Transmitters and Receivers; there is no facility for connecting network wires across pages, but Transmitters and Receivers are effectively "wires" anyway.

Appendix A.1 mentions a Usenet group (comp.graphics.apps.data-explorer). I doubt very much that group is used much by anyone. Also, the Cornell Theory Center did get rid of its gopher server some years ago as the WWW came to predominate the means for people to access the Internet. You can still connect to the DX Repository from the Theory Center Web pages.

Conclusion to Chapter 2, Appendices A.2-A.5

I hope you found some useful hints and ideas in this chapter. Scientific visualization can be very gratifying by its very nature: you get immediate feedback in a visual domain, you can make interactive changes while you study your data, you and your colleagues create new worlds for exploration and can now discuss phenomena never before seen. The net result is that scientific visualization is a powerful new medium to be employed in the discovery and experimentation stage, and provides a terrific medium to take to conferences or to mail to other scientists on videotape or CD-ROM or via the web, or by sharing the network files and data files to allow others to make their own discoveries.