Seymour Dupa wrote:
. . . . The raw data, in a
graph-dependent form?
Yes - then different graphs could be produced at a
later date if needed.
That'd work, although, as noted, the way that'd be done might be
different for different graphs. If the graph is a set of (X,Y) pairs,
the format could be a set of lines with comma-separated or tab-separated
values. If it's a set of (X,Y1,Y2,Y3,...) values (e.g., a graph with
multiple lines), that format would be generalized.
(I'm not sure I'd call that saving the *graph*, however, although
Monsieur Descartes might disagree. :-))