octave-maintainers
[Top][All Lists]
Advanced

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

Re: Compatability and an engineer's perspective


From: Jordi Gutiérrez Hermoso
Subject: Re: Compatability and an engineer's perspective
Date: Sun, 15 Jul 2012 11:28:56 -0400

On 15 July 2012 00:17, Jonathan Lister <address@hidden> wrote:
> If you want to attract MATLAB users to switch to Octave you need to make
> compatibility a top priority.

It already is. It's not fun, but we constantly keep weird finding
things that Matlab does and therefore we must do too. What gives you
the impression we don't want to make Matlab code run in Octave?

> We've looked into alternatives such as Python, Scilab, R, and
> Octave. As Octave continues down the road of making its syntax more
> and more distinct from MATLAB

What? We work really hard to avoid adding to our syntax in a way that
isn't like Matlab. What did we recently change to our syntax that
makes it impossible to run Matlab code?

> Right now I cannot recommend Octave as a replacement due to the
> differences in syntax,

What syntax works in Matlab but doesn't work in Octave? When you find
it, please file it as a bug, if it hasn't already been filed before:

    http://www.octave.org/bugs.html

> lack of handle graphics,

We've had handle graphics for about 4 years now. What makes you think
we don't have that?

> no classdef,

That's being worked on. Someone is paying for it, actually. If you
want it to happen sooner, please consider paying for it too. If you're
only paying for Matlab, how can you expect Octave to improve?

> and no IDE.

That's being worked on, also by a paid developer (Jacob Dawid):

    http://octave-gsoc2012.blogspot.com/

Again, help us secure funding for it so it can happen faster.

> The other compatibility issue I see is that the choice of FLTK for
> your widget set.

That was an old choice because it was easy to implement at the time.
We're not married to fltk. In fact, we've trying to move away from it
for some time.

> If you are not aware, all of MATLAB's UI and GUI tools are based on
> JAVA AWT and SWING. I'm afraid the choice of FLTK will limit your
> compatibility in the future.

I don't think the actual backend matters as much as long as the
uiwhatever functions are implemented. Qt is a powerful GUI toolkit
that looks a lot prettier than fltk. The problem with fltk is only
that it looks ugly, nothing else. Qt doesn't look ugly, so it can
work.

Also, you may be interested to know, someone else is paying for
bringing Java into Octave (much to our chagrin, but our chagrin can
take it if there's money involved).

> I just want the community to understand you are making it harder for
> American Engineers to switch to Octave. (IMHO)

American Engineers need to help us do this, just like anyone else.
Octave is mostly a volunteer-driven project. You get out of Octave as
much as you put into it. It's not helpful for us to know that you
looked at Octave and decided to buy Matlab licenses instead of
funnelling that money to Octave. Instead, tell us, "if you put the
money towards implementing feature foo, we can give you this much." If
you don't help us make Octave better, your criticism doesn't amount to
much.

Also, American Engineers have to understand we're not deliberately
making Octave look less like Matlab. Some things are just really hard
to implement. For example, classdef is, pardon my effusivity, a
fucking mess. I bet you don't even know how horrible it is because you
haven't used all of its (mis)features, yet, implementing classdef
means we have to copy every stupid design decision that Matlab has
about classdef. So we're not refusing to implement classdef so you can
tell us how you prefer to use Matlab instead. We haven't done it yet
because it's really difficult.

So, help us. Please.

- Jordi G. H.


reply via email to

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