Re: goals for 3.1

[Daniel J Sebald]

John W. Eaton wrote:
> On 14-Nov-2007, Joseph C. Slater PE, PhD wrote:
>
> | 
> | On Nov 13, 2007, at 2:11 PM, John W. Eaton wrote:
> | 
> | > I'd like to have a fairly small list of key goals for 3.1 so that we
> | > can make another release 6 months or so after 3.0.  Here's my current
> | > list:
> | >
> | > <snip>
> | >
> | >  * Eliminate __gnuplot_X__ functions from Octave
> | > <snip>
> | >
> | > Comments or suggestions?
> | 
> | Is there a tangible benefit to doing this?
>
> For me, there are at least two.  First, it eliminates some klugy code,
> so makes maintenance easier.  Second, since these functions no longer
> have any effect on the output of plots created with the Matlab
> compatible plotting interface, removing them will avoid some
> confusion.

I agree with that.  It is just confusion.  In fact, why not get rid of them for 
3.0?  That would give a clean break for transitioning from the old way of 
interacting with gnuplot to the new way of interacting with gnuplot.

> | Matlab compatibility. Maybe I'm wrong, but I think the existence of  
> | this level of control is a plus.
>
> Being able to specify details is nice, but tying it specifically to
> gnuplot is a big minus.
>
> Since the graphics backend may not always be based on gnuplot, I don't
> see how we can expect to support all the output formats that gnuplot
> can unless someone volunteers to write the code.

If the approach I explained in the last post were made to work, i.e., able to 
send commands to the __plot_stream__ and have whatever is on the other end 
respond accordingly, then that gives the flexibility needed.  One could write 
their own __gnuplot_X__() consisting of a dozen commands or less.  If we could 
write a script as such before 3.0 and verify it works, then post it somewhere 
that would put the issue to rest for at least the foreseeable future.

Dan