[GRASS-user] Open Source and the concept of reproducible research

[Brad Douglas]

On Thu, 2007-09-27 at 12:43 +1200, Hamish wrote:
> Michele wrote:
> > > Hi, just a little evangelism,
> > >
> > > I've just been looking at a free wavelet toolbox for Matlab
> > 
> > I don't get the point, was Matlab GPL'd lately? ;-))
> 
> 
> The WaveLab toolbox is a free download from Stanford Univ, not an
> official toolbox from the MathWorks corp. I actually found the link to it
> on an Octave website. Octave is GPL'd software which runs "in a way not
> dissimilar" to Matlab. It will run most Matlab scripts with no or only
> minor modifications.  http://www.octave.org
> 
> The issue of the Matlab engine being a black box link in the tool chain
> is raised in the reproducible research PDF article presented on the
> WaveLab website. Presumably ML can for the most part be classed in the
> same bucket as proprietary C/Fortran compilers.

I find the "reproducible research" approach very enabling, especially as
a marketing tool.  This something that GRASS lacks in many respects.
Soeren's testing suite fills some gaps, but GRASS could really use an
integrated suite to at least test modules.

Talk is cheap, but that's all I can afford at the moment. ;-)


-- 
73, de Brad KB8UYR/6 <rez touchofmadness com>