octave-dev

Re: [OctDev] Use of MATLAB Central (was Re: GPL violations)

[Jaroslav H. <hi...@gm...>]

On Mon, Aug 30, 2010 at 9:44 PM, Judd Storrs <js...@gm...> wrote:
> On Mon, Aug 30, 2010 at 3:17 PM, David Bateman <dba...@db...>
> wrote:
>>
>> Well till they make the change in the ToU I believe the question is
>> still in doubt from a legal point of view. However, if I'm ever sued,
>> I'll contact you for the version of the mail with MathWorks tech support
>> with the names to defend myself ;-)
>
> In this hypothetical situation who would be suing you? Does this
> person/entity you fear hold the copyrights to the files? If not, why would a
> court listen to their claims?
>

I suppose they couldn't sue you for copyright infringement, but for
ToS violation, stating that you caused them damage by unlawfully
exploiting the service that is intended for Matlab users and hence
limiting lawful customers by consuming server bandwidth :D
Yes, it sounds ridiculous, but wasn't it you who said that lawyers can
defend all sorts of illogical claims...?

In any case, this is good news. I'll keep my mouth shut next time...

-- 
RNDr. Jaroslav Hajek, PhD
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz

Re: [OctDev] Use of MATLAB Central (was Re: GPL violations)

[c. <car...@gm...>]

To add one point of data to this discussion,
I just noticed an interesting case that is related to
the new TMW ToS:

The NURBS toolbox was originally developed by M. Spink and distributed  
under a GPL v2 license:
http://www.aria.uklinux.net/index.php3

I made an Octave package from the NURBS toolbox some time ago and  
uploaded it on Octave-Forge
maintainig the copyright into the original files and the original  
license,
the license is clearly stated as being GPL and the copiright holder as  
being M. Spink,
you can see one example here:
http://octave.sourceforge.net/nurbs/function/vecrotx.html

I tried to contact the M.Spink at the time I started the Octave-Forge  
fork of the NURBS toolbox,
but his email address seems to be no longer active:
http://old.nabble.com/nurbs-package-ts20368543.html#a20368543

The NURBS toolbox was hosted on the Matlab Central Web Site here:
http://www.mathworks.com/matlabcentral/fileexchange/loadFile.do?objectId=312&objectType=file

but the link is now broken as the original package was removed  
probably due to the change in ToS.

I just noticed that someone re-uploaded the package on matlab central  
earlier this year:
http://www.mathworks.com/matlabcentral/fileexchange/26390-nurbs-toolbox-by-d-m-spink

this latter package seems to be identical to the original version,  
EXCEPT that
1) The copiright notice has been removed from the file headers
2) The GPL license included in the original version has been replaced  
by a file named "license.txt" containing a BSD license

I beleive this was done to comply with the new TMW ToS, but, as the  
uploader of the package is NOT the copyright holder,
This is clearly in violation of the terms of GPL.

Does any of you have suggestions about how to react to this violation?
Thanks,
c.

Re: [OctDev] Use of MATLAB Central (was Re: GPL violations)

[Judd S. <js...@gm...>]

On Sun, Sep 5, 2010 at 4:36 PM, c. <car...@gm...> wrote:
>
> I beleive this was done to comply with the new TMW ToS, but, as the
> uploader of the package is NOT the copyright holder,
> This is clearly in violation of the terms of GPL.
>

Unfortunately, my understanding is that only DM Splink (or his estate/or
someone who can demonstrate that they have legally obtained copyrights to
the work) has any actual power to enforce the copyrights to this source.
 Since none of us are copyright holders all any of us can hope to do is
contact Paul Zhang and ask for an explanation, notify DM Splink of our
suspicions, and contact Mathworks and see if they want to do something about
this.

It is not 100% clear cut that the GPL has been violated in this case
(although it seems extremely likely). It is possible that DM Splink has
distributed the source under the BSD license through other channels that we
are not aware of--for instance it is possible that Paul Zhang was able to
contacted DM Splink and obtained the source with a new license. However, the
accompanying text seems to indicate that Zhang simply changed the license to
the files because he was inconvenienced by the Mathworks change in license
policy.

If DM Splink cannot be contacted, the only other possibility would be if the
NURBS code incorporates other GPL works. This would prevent DM Splink from
distribute those components of the code under the BSD license. If that is
the case the copyright holders of the GPL components should be contacted.

This sort of mess is one of the reasons the FSF generally insists on
copyright transfer to the foundation for its works.


--judd