Seven steps for contributing new code or data to the CISM_DX package:

1. Create a stand-alone version of your contribution and name it
   CISM_DX-LastName-Keyword-YYYY.MM.DD for new codes and data.

2. Write the help documentation to conform to the style in the main
   CISM_DX package:

   a) for OpenDX macros/networks the first page should be the README
      page which contains:

      Author
      Date
      Description of network
      Explanation of Input and Output (for macros)
      When and on what operating system was the macro or network
      developed and tested

   b) for Octave m-files the header should follow the Octave conventions:
      Description of function
      Calling sequence and explanation of variables
      Author and Date
      When and on what operating system was the function developed and tested

3. Package your contribution:
   a) on Unix systems, use
      "zip -r CISM_DX-LastName-Keyword-YYYY.MM.DD.zip CISM_DX-LastName-Keyword-YYYY.MM.DD/*" and/or
      to compress and archive these directories 
      (we use .zip for Windows compatibility).
   b) on Windows systems use WinZip.

4. Email and announce it to the cismdx_user listserver:
   cismdx_user@miranda.colorado.edu

5. Upon review and testing by another CISM_DX user, some of your code
   and data from your contribution may be integrated into the next
   CISM_DX version.


Important:

1. Adjust all references and *new* data sets to the stand-alone
   directories using relative paths.  If the demo needs a data set to
   run, try to use either synthetic data or an already existing data
   in CISM_DX_DATA before submitting new data.

2. Modularize your code so the generic functions can be used by
   others.  For example, if you have a network that reads in a data
   file, transforms the data in a special way, and then generates a
   plot, remove the transform part from the network and make it a
   macro.