Re: desired features for gp backend?

[Robert T. Short]

I thought about this some after Ben's question.

I think you are right.  Even though I did like gplot, it is far better 
to abstract the plotting from the specific tool.  MATLAB plotting syntax 
is occasionally clunky, but it is reasonably straightforward.  To return 
to anything gplot specific would be going backwards, not forwards.

A user can always create his own interface if the octave solution 
doesn't work.


Plotting has always been a bit of a problem.  Back when we did 
everything in FORTRAN (just dated myself!) there were a couple of good 
subroutine packages that became fairly standardized and life was good.  
Making good plots conveniently was something I took for granted.  That 
hasn't really happened since.

Just my opinion.

Bob




John W. Eaton wrote:
> On 14-Jun-2009, Daniel J Sebald wrote:
>
> | set(fig, 'postplot', 'set gnuplotfeature xyz; replot')
> | plot(x, y)
> | 
> | might have value.
>
> I have no interest in doing this because it ties your plotting
> commands to gnuplot, and introduces portability problems for people
> who don't want to use gnuplot for plotting.  I think it is much better
> to have Octave's plotting functions and properties be generic, and
> equally supported by all plotting backends.  If you want to use
> specific gnuplot features, I think the best solution is to write your
> data to a file and use gnuplot directly.
>
> jwe
>
>
>