Re: Using MATLAB .m-files with Octave

[A+A Adler]

On Mon, 18 Oct 1999, John W. Eaton wrote:
> If the primary reason that you want to use M-files from Matlab with
> Octave is that Octave doesn't provide the features you need, then I
> believe the best long-term solution is to implement the features you
> need (independently from the Matlab sources!) and contribute them to
> Octave.

Just my two cents here:

Unfortunately, trying to duplicate Matlab M-file functionality
runs you into the typical reverse-engineering problem.

For example, I've re-implemented several of the image processing
toolbox functions (and will contribute them to octave as soon
as a get around to talking my work into letting me release them -
since I wrote them at work)

To implement the general functionality isn't hard - for example
you can write a sobel edge filter after reading any decent textbook.

However, if you want full compatibility with Matlab then there
is no other way than by looking in the M file to see what
threshold values they use. 

There is no way to do "clean room" reverse engineering in
this case.

Unfortunately, as I understand it, the law is on this subject
extremely complicated and varies among jurisdictions as to the
legality of this.

Does anyone understand this issue better than I?
______________________________________________________________
Andy Adler,                               address@hidden



---------------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.  To ensure
that development continues, see www.che.wisc.edu/octave/giftform.html
Instructions for unsubscribing: www.che.wisc.edu/octave/archive.html
---------------------------------------------------------------------