On 06/08/11 04:50, Ethan Merritt wrote:
> On Tuesday, June 07, 2011 06:05:40 pm Clark Gaylord wrote:
>> I have always felt the gnuplot.info name should be canonical. It points
> to sf for v4 and vt for v6 and is much more portable.
> But what is it that one is citing, exactly?
> The source is not (so far as I know) available on gnuplot.info,
> So if the idea is to cite the location of the program source, then *·info
> may not be the correct URL.
> If the idea is to cite the manual, say, then that's another issue.
> Perhaps we should register the manual with a repository that would
> issue a permanent DOI. arXiv.org?
What is the aim here? The fact that you are having to do contortions to
fit this into the reference citation format for the publications
suggests you are misusing it. That may not go down well with the publisher.
While it is nice to recognise the software in a paper using it , it is
after all only a plotting tool and there is no reason to cite it as a
reference since it should not affect the results in any way.
The only exception I can think of is if your are using it to do linear
regression or something in which case you should probably understand how
it does that and comment on what methods *you* have adopted to fit
straight lines , functions, whatever, in choosing to do that with a
particular tool that automates the process.
To recognise gnuplot and help the reader reproduce your work it would be
sufficient to name gnuplot as an open source plotting tool. Anyone
capable of using it should be able to find without further help.
Mentioning gnuplot as the tool used to produce any graphics seems like a
good idea. I doubt citing it as a reference is appropriate.