help-octave
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GUI frontend for Octave


From: John W. Eaton
Subject: Re: GUI frontend for Octave
Date: Mon, 12 Nov 2007 13:51:11 -0500

On 12-Nov-2007, Kamaraju Kusumanchi wrote:

| Another thing that comes to my mind is this. There is a package called 
| texmacs which provides a GUI for octave 2.1. However it does not work 
| with octave 2.9. Upon searching further, I found out http://
| www.texmacs.org/tmweb/plugins/numerics.en.html#octave where it says
| 
| > A first experimental interface between TeXmacs and GNU Octave 
| > has been written by Michael Graffam. Unfortunately, we get little
| > help from the main Octave  developers for the improvement of the
| > interface and the integration of patches into the mainline of Octave.
| > So, if you like the idea of using Octave from inside TeXmacs, then
| > please help advertising it.

I'm not sure what to make of this.  I don't recall what might have
happened that would have lead to the conclusion that the "main Octave
developers" didn't provide enough help, or what patches were submitted
or why they might have been rejected (maybe there were good reasons?).

| Based on this, If I am the manager I wouldn't put money into making a GUI 
| unless the project is approved by the lead developers of octave. AFAIK, 
| money is not the issue here. It is how much the lead developers of octave 
| want to have a GUI...
| 
| You can say it is GPL, you can write your own code etc., I understand all 
| that. If I desperately want a GUI, I probably will do that. But I have a 
| feeling that money is not the stumbling block here.

A number people have come along and said that they are going to write
a GUI for Octave, and that they are experts and so to them, GUIs are
easy, it will only take a week, etc., etc.  I think in all cases so
far, the people doing this have thought of their GUI/IDE project as
separate from Octave and not needing to be "approved by the lead
developers of Octave".  Then they typically have gotten something sort
of working, then (more or less) abandoned the project.

Although it might be possible to write a nice package that provides
bindings to a GUI toolkit, an effective IDE probably needs closer
integration with the Octave core.  I think the key problem is how to
integrate Octave's command line with the IDE.  Who should be in
charge, the IDE or Octave?  To get this right will probably require
some modifications to Octave so that it can properly handle input and
events.  To make interaction with the Octave prompt as functional in
the IDE as it is in the typically shell window, the IDE will either
have to provide a complete reimplementation of readline, or provide a
terminal window capable of running readline.  Doing all of this in a
portable way makes it even more challenging.

In any case, I have no objection to GUI stuff (either scriptable GUI
elements or an IDE) for Octave, but I don't think it helps to have N
half-baked and semi-abandoned projects lying around.  I think it would
be better if the people interested in these things would work together
with us instead of splitting their effort among many separate
projects.

So, If you are interested in working on an IDE for Octave, then please
join the maintainers list and discuss your ideas there.  But before
posting, maybe search the archives and review some of the previous
discussions.

jwe


reply via email to

[Prev in Thread] Current Thread [Next in Thread]