Using a standard file format in a conventional file-oriented computing environment, it is convenient to use a standard naming convention for the data files written in that format.
It should be noted that this naming convention is NOT a part of the proposed standard data file format, but instead co-exists alongside it as a separate entity which others may find useful.
Different systems, of course, have wildly different constraints on file names--lengths, extensions, legal characters, etc. The naming convention used should be adaptable for use with as many systems as possible.
The basic plan adopted here is to compose a file name of text fields separated if possible by delimiters. The first field should signify the physical quantity the file contains, and the last should indicate where the data came from. Each of the rest of the fields, included only if there is room and if the dataset creator desires, begins with a specific character distinguishing it from the others. (This allows for automated parsing of file names.)