[SciPy-user] scipy code contribution question
Thu Jul 5 07:36:28 CDT 2007
I certainly don't claim to be an expert on derivative work, but it
seems like you aren't creating a derivative work if Mathworks never
showed you the source code - which I assume they didn't.
It seems like an overly literal interpretation of the license snippet
that Bill posted would say that if I ever used Matlab, then I can
never write code that does the same things as Matlab or any of its
toolboxes. If that is legally defended, then I can't contribute
anything to scipy.
I appreciate that we want to be careful, but it doesn't make sense
that we could get in trouble by creating functions with the same names
as Matlab functions if we've never seen their source code. I mean,
can they copyright plot(x,y)? If so, than matplotlib is in trouble
(unless John had never used Matlab when he came up with that syntax).
What about fft(x)?
The link that Bill posted with the GPL code poses as big a problem in
my mind. If any of us looks at that code, then any similar algorithm
would need to be GPL'ed as a derivative work. So, don't open that
On 7/5/07, Bill Baxter <email@example.com> wrote:
> On 7/5/07, Bill Baxter <firstname.lastname@example.org> wrote:
> > On 7/5/07, David Cournapeau <email@example.com> wrote:
> > > Lev Givon wrote:
> > > >
> > > As everybody here I guess, IANAL, but the safest thing to do is to
> > > explicitely ask the mathworks.
> > The license I got with my copy of Matlab is pretty clear:
> > """
> > Except as expressly provided by this Agreement, including the
> > attached Addendum, Licensee may not adapt, translate, or convert
> > "M-files", "MDL-files" or "P-code" contained in the Programs in
> > order to create software, a principal purpose of which is to
> > perform the same or similar functions as Programs licensed by
> > MathWorks or which is intended to replace any component of the
> > Programs. The Licensee may not incorporate or use "M-files",
> > "P-code", source code, or any other part of the Programs in or
> > as part of another computer program without the consent of
> > MathWorks.
> > """
> > You can contact them and try, but I don't expect they'd be willing to give
> you explicit consent to translate their M-files in this case, since is
> precisely for the purpose of creating software that does the same thing as
> Matlab. And even worse it's for creating *free* software that does the same
> thing as Matlab.
> > On the other hand, if Octave has an implementation of this particular
> function you're after, then you are much more likely to get the developers
> permission to translate it to Python and put it under BSD.
> And here appears to be just such a beast:
> SciPy-user mailing list
More information about the SciPy-user