Plotting using R?



--- Daniel Lakeland <dlakelan@street-artists.org> wrote:

> Talking to R by dumping data files and sending commands down a 
> pipe to the interpreter is certainly not less klugey than
> doing the same for gnuplot, but R is actively being developed and 
> would be a fantastic coupling.

How tied to R itself are the plotting routines?  I don't suppose they
could be stripped out and used as a stand alone setup?

In any case, I can see the attraction of being able to make use of the
R plotting system if it is already installed on the system.

> http://www.r-project.org for those who don't know about R.
> 
> I have considerable experience with R plotting, and have done a fair
> amount of common lisp coding (though not in the last 2 years), so if
> someone wants to take me under their wing a little and give me hints
> of where to start, I'd be happy to do a proof of concept maxima
> package that does 2D plotting.

Well, that wouldn't be me, but at a guess the best place to start is
probably to look at and compare/contrast the openmath/gnuplot/geomview
output routines.  As far as I know in all three cases a file called
maxout.*program name* is produced and used, so you might be able to
adapt one of those routines to produce R notation and call R.  It is a
kludgy way of doing it regardless of the program used to display it,
but  it's simple and effective. I expect it's one of those things where
a solution that is 5x better/more elegant is probably 20x the work, and
so will have to wait :-(.  It would be very interesting to see what R
can do - IIRC I have R installed on my machine at home and I would be
glad to help in testing.

CY


		
_______________________________
Do you Yahoo!?
Declare Yourself - Register online to vote today!
http://vote.yahoo.com